Teradata Release 13.11 Viewpoint User Guide
Teradata Viewpoint is a powerful tool for managing your Teradata Database system. It provides a centralized portal for monitoring system health, performance, and capacity. With a variety of portlets, you can easily view key metrics, track queries, and manage your system resources.
advertisement
Teradata Viewpoint
User Guide
Release 13.11
B035-2206-120A
December 2010
The product or products described in this book are licensed products of Teradata Corporation or its affiliates.
Teradata, BYNET, DBC/1012, DecisionCast, DecisionFlow, DecisionPoint, Eye logo design, InfoWise, Meta Warehouse, MyCommerce,
SeeChain, SeeCommerce, SeeRisk, Teradata Decision Experts, Teradata Source Experts, WebAnalyst, and You’ve Never Seen Your Business
Like This Before are trademarks or registered trademarks of Teradata Corporation or its affiliates.
Adaptec and SCSISelect are trademarks or registered trademarks of Adaptec, Inc.
AMD Opteron and Opteron are trademarks of Advanced Micro Devices, Inc.
BakBone and NetVault are trademarks or registered trademarks of BakBone Software, Inc.
EMC, PowerPath, SRDF, and Symmetrix are registered trademarks of EMC Corporation.
GoldenGate is a trademark of GoldenGate Software, Inc.
Hewlett-Packard and HP are registered trademarks of Hewlett-Packard Company.
Intel, Pentium, and XEON are registered trademarks of Intel Corporation.
IBM, CICS, RACF, Tivoli, and z/OS are registered trademarks of International Business Machines Corporation.
Linux is a registered trademark of Linus Torvalds.
LSI and Engenio are registered trademarks of LSI Corporation.
Microsoft, Active Directory, Windows, Windows NT, and Windows Server are registered trademarks of Microsoft Corporation in the
United States and other countries.
Novell and SUSE are registered trademarks of Novell, Inc., in the United States and other countries.
QLogic and SANbox are trademarks or registered trademarks of QLogic Corporation.
SAS and SAS/C are trademarks or registered trademarks of SAS Institute Inc.
SPARC is a registered trademark of SPARC International, Inc.
Sun Microsystems, Solaris, Sun, and Sun Java are trademarks or registered trademarks of Sun Microsystems, Inc., in the United States and other countries.
Symantec, NetBackup, and VERITAS are trademarks or registered trademarks of Symantec Corporation or its affiliates in the United States and other countries.
Unicode is a collective membership mark and a service mark of Unicode, Inc.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Other product and company names mentioned herein may be the trademarks of their respective owners.
The information contained in this document is provided on an “as-is” basis, without warranty of any kind, either express or implied, including the implied warranties of merchantability, fitness for a particular purpose, or non-infringement.
Some jurisdictions do not allow the exclusion of implied warranties, so the above exclusion may not apply to you. In no event will Teradata Corporation be liable for any indirect, direct, special, incidental, or consequential damages, including lost profits or lost savings, even if expressly advised of the possibility of such damages.
The information contained in this document may contain references or cross-references to features, functions, products, or services that are not announced or available in your country. Such references do not imply that Teradata Corporation intends to announce such features, functions, products, or services in your country. Please consult your local Teradata Corporation representative for those features, functions, products, or services available in your country.
Information contained in this document may contain technical inaccuracies or typographical errors. Information may be changed or updated without notice. Teradata Corporation may also make improvements or changes in the products or services described in this information at any time without notice.
To maintain the quality of our products and services, we would like your comments on the accuracy, clarity, organization, and value of this document. Please e-mail: [email protected]
Any comments or materials (collectively referred to as “Feedback”) sent to Teradata Corporation will be deemed non-confidential. Teradata
Corporation will have no obligation of any kind with respect to Feedback and will be free to use, reproduce, disclose, exhibit, display, transform, create derivative works of, and distribute the Feedback and derivative works thereof without limitation on a royalty-free basis.
Further, Teradata Corporation will be free to use any ideas, concepts, know-how, or techniques contained in such Feedback for any purpose whatsoever, including developing, manufacturing, or marketing products or services incorporating Feedback.
Copyright
©
2008 - 2010 by Teradata Corporation. All Rights Reserved.
Table of Contents
Chapter 1:
Teradata Viewpoint Help
Chapter 2:
Getting Started
Chapter 3:
Portal Basics
Chapter 4:
Portlet Basics
Chapter 5:
Portal and Portlet Tools
Teradata Viewpoint User Guide, Release 13.11
3
Table of Contents
Chapter 6:
Profile
4
Chapter 7:
System Administration
Chapter 8:
Alert Viewer
Teradata Viewpoint User Guide, Release 13.11
Table of Contents
Chapter 9:
Calendar
Chapter 10:
Canary Response Times
Chapter 11:
Capacity Heatmap
Chapter 12:
External Content
Chapter 13:
Lock Viewer
Teradata Viewpoint User Guide, Release 13.11
5
Table of Contents
6
Chapter 14:
Metrics Analysis
Chapter 15:
Metrics Graph
Chapter 16:
My Queries
Chapter 17:
Node Resources
Chapter 18:
Productivity
Teradata Viewpoint User Guide, Release 13.11
Table of Contents
Chapter 19:
Query Monitor
Chapter 20:
Query Spotlight
Chapter 21:
Remote Console
Chapter 22:
Space Usage
Teradata Viewpoint User Guide, Release 13.11
7
Table of Contents
Chapter 23:
SQL Scratchpad
Chapter 24:
System Health
Chapter 25:
Today's Statistics
Chapter 26:
TVS Monitor
8 Teradata Viewpoint User Guide, Release 13.11
Table of Contents
Chapter 27:
Viewpoint Monitoring
Chapter 28:
Workload Designer
Chapter 29:
Workload Health
Chapter 30:
Workload Monitor
Teradata Viewpoint User Guide, Release 13.11
9
Table of Contents
10 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 1
Teradata Viewpoint Help
Teradata Viewpoint Help
The Teradata Viewpoint Help online user guide provides targeted information about how to use the Teradata Viewpoint portal and related system management tools, called portlets, to manage your Teradata Database system.
Click to open a browser containing Teradata Viewpoint Help. The display is divided into panes:
Contents
Lists the hierarchy of topics contained in Teradata Viewpoint Help. To navigate using this pane:
• Click next to a topic name to expand the list of available topics.
• Click next to a topic name to collapse the list of available topics.
• Click on any topic name to display corresponding topic information in the right pane.
Current Help Topic
Displays the currently selected Help topic. In this pane, you can:
• Browse through topic information, using the Scroll Bar to view content not visible in the pane.
• Select links to related topics that are above (parent) or below (child) the current topic in the content hierarchy.
Only one instance of the Help system can be displayed regardless of how many sessions are open.
Teradata Viewpoint User Guide, Release 13.11
11
Chapter 1 Teradata Viewpoint Help
12 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 2
Getting Started
Getting Started
Getting Started topics describe how to begin and end a Teradata Viewpoint session.
Logging On
Logging on to the Teradata Viewpoint portal begins your session so you can begin working with the Teradata Viewpoint portal.
1
2
3
Open a browser.
Enter the address for your Teradata Viewpoint portal.
The Welcome page appears, with the portal version number shown at the bottom.
Log on to the Teradata Viewpoint portal.
If your Teradata Viewpoint system is set up to create a user profile automatically, the username and password you enter are authenticated against your company-provided username and password the first time you log on to Teradata Viewpoint. Automatic profile creation is known as auto-provisioning.
Logging Off
Logging off the Teradata Viewpoint portal ends your current session and returns you to the
Welcome page.
1
From the Teradata Viewpoint main menu bar, click Logout to exit the portal.
All page content is preserved for your next Teradata Viewpoint session. Any open
Teradata Viewpoint Help window is closed.
Teradata Viewpoint User Guide, Release 13.11
13
Chapter 2 Getting Started
14 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 3
Portal Basics
Portal Basics
To help you work efficiently, Teradata Viewpoint uses a page metaphor as the framework for displaying and updating portlets. Each portal page is a virtual work space where you decide which portlets to display and how to arrange them on the page. Examples of ways to organize your work include defining a page for each system being monitored, or for each type of query or user. As you work, Teradata Viewpoint continually updates the information displayed on the page that currently fills the Teradata Viewpoint portal. This page is called the active page.
Manage portal pages using the following guidelines:
• Add portal pages at any time during a Teradata Viewpoint session.
• Access any portal page by clicking its tab; only one page can be active at a time.
• Change the name of any tab, including the Home page tab; page names can be duplicated.
• Rearrange pages by dragging and dropping into a new location.
• Remove pages, along with any portlets contained on the page, with a single mouse-click.
• One page (tab) must remain, as well as the Add Page tab.
Adding a Portal Page
Organize your system management tools by adding pages to the Teradata Viewpoint portal.
Multiple pages can be added or removed per session. The newest page is always the active page unless you click on another tab.
1
In the Teradata Viewpoint portal, click Add Page.
A New Page tab appears to the left of Add Page and becomes the active page.
Naming a Portal Page
One way you can customize the Teradata Viewpoint portal is by giving meaningful names to the pages that display portlets. Initially, each page is named New Page. You can change the page name at any time during a Teradata Viewpoint session.
Teradata Viewpoint User Guide, Release 13.11
15
Chapter 3 Portal Basics
1
2
3
4
Click on the page tab to highlight the page name.
Enter the new name.
Page names can be up to 30 characters long.
To apply the new name, do one of the following:
• Click anywhere on the portal page.
• Press Enter.
[Optional] Reorder the portal pages to maintain your management scheme.
Arranging Portal Pages
Initially portal pages appear in the order in which they have been added from left to right.
Arranging or rearranging pages does not change page content, and can be done at any time during your Teradata Viewpoint session.
1
In the portal task bar, mouse over the tab you intend to move.
The cursor changes to a directional cursor.
2
Drag the tab to the left or to the right of its current position.
The tab aligns in the new position.
Add Page is always in the right-most position and cannot be moved or deleted.
Deleting a Portal Page
A portal page and its contents can be deleted from the portal at any time after it has been defined and at any time during your Teradata Viewpoint session.
1
In the portal task bar, click the tab of the page you intend to delete.
2
3
Click to remove the portal page and its contents.
A confirmation message appears.
Click OK to delete the page.
The page is removed, and the remaining tabs shift left.
16 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 4
Portlet Basics
Portlet Basics
Teradata Viewpoint system management tools are called portlets. Select the portlets that you want to display and monitor from submenus, or categories, under Add Content. You can also search for a specific portlet name using the filter feature. Generally, every instance of a portlet:
• Has a frame that appears when the cursor moves over any part of the portlet on the page and disappears when the cursor moves off the portlet.
• Displays the portlet name or the current settings in the upper frame, depending on the current activity.
• Has a width requirement to ensure proper display of its graphical information such as charts, sparklines, or graphs.
• Remains at a fixed size even when the browser window is re-sized. You can use the browser scroll bars to view the entire portlet view.
• Can be repositioned on the portal page.
Portlet Controls
The Teradata Viewpoint Administrator controls what portlets and features you are authorized to use. If a portlet or feature is not available, the Teradata Viewpoint
Administrator can assign permissions. The following controls appear within each portlet frame if the control is available:
Teradata Viewpoint User Guide, Release 13.11
17
Chapter 4 Portlet Basics
18
Back
Returns to the previous view.
Rewind
Indicates that the portlet can be set to display data from a previous point in time.
Preferences
Accesses portlet preferences and settings. Preferences are used to specify what information is displayed, time intervals for reporting, and other features that help you customize the portlet functions.
Share Portlet
Captures a customized version of a portlet for use by other users. The Teradata
Viewpoint Administrator must make the customized portlet available for sharing.
Collapse
Toggles the portlet closed. This button appears only when the portlet is open. When collapsed, only the upper and lower portlet frame sections are displayed.
Expand
Toggles the portlet open. This button appears only when the portlet is collapsed.
When expanded, the portlet returns to its normal size and position on the portal page.
Maximize
Toggles the portlet to fill the portal page, covering all other portlets being displayed.
Minimize
Toggles the portlet to the default size of one column wide.
Remove
Removes the portlet and all its settings from the active portal page.
Adjust Height
Adjusts the vertical size of the portlet. Drag the control to resize the portlet vertically. The portlet retains its new dimensions until you click the Restore Height button.
Restore Height
Restores the portlet to its default height. This button appears only when you change the default height of the portlet.
Each portlet instance has its own settings and controls as well as the features shared by all portlets. Depending on the specialized nature of the portlet, a portlet can include any of the following features at the top:
• Tabs for accessing views, when views are available to organize settings according to function.
• Selection menus above the toolbar to choose detailed information.
• A toolbar with buttons for quick access to information.
Teradata Viewpoint User Guide, Release 13.11
Chapter 4 Portlet Basics
• Date-range slider for selecting start dates and end dates for analysis.
• Information balloons to show detailed information when you mouse over the portlet.
On the bottom bar, some portlets also feature controls for adjusting and restoring height.
Adding a Portlet
A portlet can be added to the active page at any time during a session. Except for the
Calendar portlet, multiple instances of a portlet can be added to a portal page.
1
2
3
In the portal, click Add Content.
Select a portlet from the submenus.
The portlet is added at the top of the left-most column.
[Optional] Reposition the portlet on the portal page by dragging it to another location.
Filtering Portlets
Find a portlet by using the keyword filter box and the SEARCH RESULTS list. This method is also useful when you want to add a portlet to the active page and the list of portlets is long.
1
2
3
Click in the keyword filter box to activate the field for typing.
Begin typing the portlet name.
As you type, the SEARCH RESULTS list shows the portlets that match your filter criteria. This field is not case-sensitive.
Click a portlet name in the SEARCH RESULTS list to add it to the page.
Deleting a Portlet
A portlet can be deleted from the active portal page at any time. Deleting a portlet resets portlet information and deletes any custom settings set for that instance only. Other portlet instances on the same page or on other pages are not affected.
1
On the active portal page, mouse over the portlet frame to activate the portlet.
The portlet does not need to be expanded if it is collapsed.
Teradata Viewpoint User Guide, Release 13.11
19
Chapter 4 Portlet Basics
2
3
Click to remove the portlet from the portal page.
A confirmation message appears.
Click OK.
The portlet is removed from the active portal page.
20 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 5
Portal and Portlet Tools
Portal and Portlet Tools
Portal and Portlet Tools topics describe the tools that control the functions accessed from the portal and portlet frames. These tools are generally available to all users and allow you to customize the Teradata Viewpoint portal and portlets to make managing and working with system data more efficient. Use these functions to customize Teradata Viewpoint and streamline your workflow:
• Rewind
• Pause
• Share Portlet
About the Rewind Feature
The rewind feature allows you to view data that corresponds to dates and times in the past and compare it to data for a different date and time. You can rewind the data for some or all portlets on a portal page to a previous point in time, such as when a job failed. Rewinding portlet data is useful for identifying and resolving issues.
You can rewind data as far back as data is available. The rewind feature is not available for portlets that have portlet-specific methods for reviewing data over time.
Using the rewind toolbar, you can enter a specific date and time as well as scroll through the data in increments of seconds, minutes, hours, or days. All portlets on the page that are participating in rewind activities display data that corresponds to the selected rewind date and time each time a selection is made.
If portlet data can be rewound, a time stamp, , and appear in the portlet frame.
You can also do the following:
• Pause individual portlets so that the data display is no longer rewound or refreshed
Teradata Viewpoint User Guide, Release 13.11
21
Chapter 5 Portal and Portlet Tools
• Resume participation in rewind for a portlet in pause mode
• Prevent a portlet from participating in rewind and display current data instead
• Resume participation in rewind and return to the state a portlet was in when it was unlinked from rewind
• Cancel the rewind feature so that all rewound portlets on the portal page display current data
Permission to use the rewind feature is granted to individual users based on their role and portlet access. If the rewind toolbar does not appear on the portal page, you are not authorized to use the feature. If the rewind icons and time stamp do not appear in the portlet frame, the feature is not available for that portlet.
Rewinding a Portal Page
You can rewind all participating portlets on a portal page. Use the rewind feature to:
• Select a date and time for data to display in all participating portlets currently on a portal page or added to the page.
• Scroll the selected time incrementally to find a change in status among the participating portlets displayed on the page.
1
2
3
4
Click .
The Rewind toolbar appears along the top of the portal page.
[Optional] Specify the date and time:
a
Enter a date in the Desired Date and Time date field.
b
Enter a time in the Desired Date and Time time field.
c
Click anywhere in the portal page to display the data that corresponds to the selected date and time for all participating portlets.
[Optional] Drag left to scroll back through time in increments of:
• 1 second
• 10 seconds
• 1 minute
• 10 minutes
• 1 hour
• 1 day
[Optional] Drag right to scroll forward through time in increments of:
• 1 second
• 10 seconds
• 1 minute
• 10 minutes
• 1 hour
• 1 day
You cannot scroll forward past the current time as projected data is not available using the rewind feature.
22 Teradata Viewpoint User Guide, Release 13.11
Chapter 5 Portal and Portlet Tools
About Query Spotlight and Rewind
The Query Spotlight portlet cannot rewind and does not use the rewind toolbar. The
Control Rewind button in the Query Spotlight portlet is a toggle that enables portal rewind for portlets that support the rewind function on the same page. When rewind is enabled, the time marker in Query Spotlight sets the time and date in the rewind toolbar and controls the time and date for data in the other portlets. This allows you to investigate how metrics compare across portlets during the same time frame.
When Control Rewind is toggled off, the time marker no longer controls the date and time for data in the other portlets and portal rewind is controlled from the rewind toolbar.
Pausing Rewind for a Portlet
You can pause an individual portlet on a portal page while other portlets on the page display rewound data. Pausing an instance of a portlet freezes the display of data at the moment the portlet is placed in pause mode, allowing you to compare data for multiple rewound time periods.
1
2
Click in the portlet frame.
Click Pause.
Data displayed in the portlet becomes static and does not change until you resume the portlet refresh rate.
Resuming Rewind for a Portlet
You can resume using the rewind feature for an individual portlet on a portal page that is in pause mode.
1
2
Click in the portlet frame.
Click Resume.
Data displayed in the portlet is updated to the selected rewind time and does not change until the selected date or time for the rewind feature is changed.
Stopping Rewind Participation
You can prevent an individual portlet from participating in rewind when the portlet frame displays .
1
2
Click in the portlet frame.
The portlet can be in pause mode or actively responding to the rewind date and time controls.
Click Unlink from Rewind.
The portlet no longer responds to rewind date or time controls and is updated to display the most recent data available.
Restarting Rewind Participation
You can have an individual portlet not participating in rewind resume participation when the portlet frame displays .
Teradata Viewpoint User Guide, Release 13.11
23
Chapter 5 Portal and Portlet Tools
1
2
Click in the portlet frame.
Click Link to Rewind.
The portlet returns to the mode it was in when it was unlinked from participation in rewind. For example, if the portlet was in pause mode when unlinked from rewind, it is in pause mode when participation in rewind resumes.
Canceling the Rewind Feature
You can turn off the rewind feature and resume display of current data in all portlets on the portal page.
1
Click Cancel.
All portlets on the portal page that were participating in rewind activities end participation and display the most recent data available.
About the Pause Feature
Any rewindable portlet can be paused from refreshing data. Pausing an instance of a portlet freezes the display of data at the moment the portlet is placed in pause mode, allowing you to compare data for different time periods. The rewind feature does not need to be active to pause the portlet. When a portlet is in pause mode, the time stamp flashes, indicating that the data in the portlet is no longer being refreshed.
Pausing Portlet Refresh
You can pause an individual portlet on a portal page.
1
2
Click in the portlet frame.
Click Pause.
Data displayed in the portlet becomes static and does not change until you resume the portlet refresh rate.
Restarting Portlet Refresh
You can resume the portlet refresh rate when the portlet is in pause mode.
1
2
Click in the portlet frame.
To restart the portlet, do one of the following:
• Click Refresh to refresh portlet summary and detail data to the current date and time.
• Click Resume to resume the portlet refresh rate.
About Shared Portlets
Use the Share Portlet feature to make a customized or preconfigured version of a Teradata
Viewpoint portlet available to other users. When the shared portlet is added to a portal page, all preferences and settings you selected when you customized the portlet are used.
Any user can customize a shareable portlet and submit it for sharing. Depending on your
Teradata Viewpoint system configuration, the portlet is not available to authorized users
24 Teradata Viewpoint User Guide, Release 13.11
Chapter 5 Portal and Portlet Tools until the Teradata Viewpoint Administrator approves the request. Only the Teradata
Viewpoint Administrator can delete the shared portlet entirely. If deleted, existing instances remain on your portal page. However, a warning message appears if you try to access the portlet.
The Shared Portlets feature allows you to:
• Share customized views of a Teradata Viewpoint portlet with other users.
• Create different configurations of a portlet for different roles.
• Assign a preconfigured portlet to a role as a default instance of the portlet.
• Assign a name to the preconfigured portlet.
The portlet name appears in the Shared submenu of the Add Content menu. The Shared submenu only appears when at least one shared portlet is enabled for the user who is logged on.
Creating a Shared Portlet
Use the Share Portlet feature to customize a Teradata Viewpoint portlet and make it available to other users.
1
Add a portlet instance that you want to share to your portal page.
2
3
4
5
6
7
8
From the portlet frame, click to access the PREFERENCES view.
Customize the portlet using the available preference settings.
Click OK to save the settings.
Click to access the SHARE PORTLET dialog box.
Enter a name for the custom portlet using up to 25 alphanumeric characters.
Depending on your Teradata Viewpoint system configuration, the portlet is not available to authorized users until the Teradata Viewpoint Administrator approves your request.
Click Save.
[Optional] Contact the Teradata Viewpoint Administrator to verify that the portlet has been received for approval.
Teradata Viewpoint User Guide, Release 13.11
25
Chapter 5 Portal and Portlet Tools
26 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 6
Profile
Profile
The Profile portlet provides tabs that allow you to set your personal user settings and manage your Teradata Database profile information:
Personal Info
Change your name, change your password (if authorized), and set Teradata
Viewpoint preferences.
Teradata Profiles
Manage settings for each of your Teradata Database user profiles. Link your
Teradata profile to the Teradata Database systems you are authorized to access.
Access the Profile portlet from the Teradata Viewpoint portal main menu bar.
About the Personal Info Tab
The Profile view, Personal Info tab, allows you to manage your Teradata Viewpoint profile:
• Change the first and last name for the profile.
• Change your Email address.
• Change your Locale.
• Select the Time Zone.
Use the Personal Info tab to set your locale. Supported locales include:
• Chinese (China and Taiwan)
• English (US and UK)
• French
• German
• Japanese
• Korean
The locale you select is applied to the entire Teradata Viewpoint session and is maintained in future sessions.
The list of available locales is determined by the language packs installed. If the locale you want does not appear in the list, contact the Teradata Viewpoint Administrator.
Changing a Name
You can change the first and last name for a Teradata Viewpoint profile.
Teradata Viewpoint User Guide, Release 13.11
27
Chapter 6 Profile
28
1
2
3
4
From the Profile view, click the Personal Info tab.
Change the name in the First Name field.
Change the name in the Last Name field.
Click Apply.
A confirmation message appears.
Setting a Contact Email Address
You can set your Email address using the Personal Info tab. This email address is used for identification and communication in the Teradata Viewpoint system, including notifications and reminders.
The email address cannot contain uppercase letters. Uppercase characters are automatically converted to lowercase.
1
2
3
From the Profile view, click the Personal Info tab.
Enter your full address in the Email field.
Click Apply.
A confirmation message appears.
Managing Passwords
You can manage Teradata Viewpoint profile passwords. A password must be 4 to 25 characters in length.
A user cannot change their password if their profile is externally authenticated. The fields required to change the password are not available when external authentication is used. To change the password on an externally authenticated profile, follow your company policy for password management.
1
2
3
4
5
From the Profile view, click the Personal Info tab.
Enter your current password in the Current Password field.
Enter a new password in the New Password field.
Confirm your new password by retyping it in the Re-enter New Password field.
Click Apply.
A confirmation message appears.
Setting the Locale
You can set the Teradata Viewpoint locale to display the local language and dates, times, and numbers in the correct format for the country.
1
2
3
From the Profile view, click the Personal Info tab.
Select the locale from the list.
The Teradata Viewpoint application server locale is the default.
Click Apply.
Teradata Viewpoint User Guide, Release 13.11
Chapter 6 Profile
A confirmation message appears.
Setting the Time Zone
You can adjust your local Teradata Viewpoint time zone setting to compensate for different time zones. Select the local time zone to display all time and date information in the local date and time, even if the Teradata Database system you are monitoring is located in a different time zone.
The time zone you enter compensates for the time difference between the monitored system location and the Teradata Viewpoint reporting location.
1
2
3
From the Profile view, click the Personal Info tab.
Select a time zone from the list.
Time zones shown are relative to Greenwich Mean Time (GMT) and include a reference location. The Teradata Viewpoint application server time zone is the default. To view more time zones, use the scroll bar.
Click Apply.
A confirmation message appears.
About the Teradata Accounts Tab
Use the Teradata Accounts tab in the Profile view to manage Teradata Database account logon and profile information for each Teradata Database system. Teradata account information is used by the My Queries portlet. For each Teradata Database system, you can:
• Add accounts
• Edit existing profiles
• Delete profiles
System profiles added using the Teradata Accounts tab must already exist on the monitored system. The system validates the USERNAME, PASSWORD, and ACCOUNT STRING fields prior to confirmation.
Adding a User Profile
This task explains how to associate Teradata Database profiles with your Teradata
Viewpoint profile.
1
2
3
4
From the Profile view, click the Teradata Accounts tab.
Under Add Account, select a Teradata Database system from the list.
Enter the following information in the appropriate fields:
• Username
• Password
• [Optional] Account string
Click Add.
A confirmation message appears.
Teradata Viewpoint User Guide, Release 13.11
29
Chapter 6 Profile
Editing a User Profile
This topic explains how to edit a Teradata Database system user profile in Teradata Viewpoint.
1
2
3
4
5
6
From the Profile view, click the Teradata Accounts tab.
Mouse over the row containing the user profile information for the Teradata Database system you are updating.
The row is highlighted, and Edit and Delete appear.
Click Edit to the right of the row containing the user profile information.
Edit any of the following fields:
•
PASSWORD
•
ACCOUNT STRING
[Optional] Click Test to verify that the logon settings are correct.
A confirmation message appears.
Click Save to save your changes.
A confirmation message appears.
Deleting a User Profile
This task explains how to delete a Teradata Database system user profile from Teradata
Viewpoint using the Profile view, Teradata Accounts tab.
1
2
3
From the Profile view, click the Teradata Accounts tab.
Mouse over the row containing the Teradata Database user profile information for the system you are updating.
The row is highlighted, and Edit and Delete appear.
Click Delete to the right of the row containing the user profile information.
If the deletion is successful, a confirmation message appears.
30 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 7
System Administration
System Administration
The Teradata Viewpoint administrative portlets allow the Teradata Viewpoint
Administrator to provide access to Teradata Viewpoint resources and information.
Configuration portlets
Teradata Systems Configure, enable, and disable Teradata Viewpoint servers and data collectors. After a server is defined to Teradata Viewpoint, you can maintain logins, accounts, passwords, and character set settings.
Managed Servers Configure, enable, and disable managed servers.
Alert Setup Configure alerts or migrate Teradata Manager alerts.
LDAP Servers Configure LDAP servers to authenticate, auto-provision, and map
Viewpoint roles to Teradata Viewpoint users.
Backup Configure daily backups.
Portlet Library
Select portlets for activation. Using a simple checklist, you can either enable or restrict access to available Teradata Viewpoint portlets.
Roles Manager
Manage roles, assign users, and grant permissions efficiently. After a role is created, you can customize the role by assigning users, enabling portlets, grant permissions for metrics, and granting permissions to roles for portlets.
User Manager
Manage Teradata Viewpoint user accounts by creating user accounts, assigning or resetting passwords, and assigning users to predefined roles.
About Security and Permissions
The Teradata Viewpoint administrative portlets allow the Teradata Viewpoint
Administrator to control portlet access at different levels by setting permissions for Teradata
Viewpoint users.
Global
Enable or disable access to portlets globally using the Portlet Library portlet. This setting takes precedence over every other portlet permission level.
Teradata Viewpoint User Guide, Release 13.11
31
Chapter 7 System Administration
Role
Assign permissions to classes of users called roles using the Roles Manager portlet.
Each tab controls access to features.
User
Assign available roles to users and set role precedence using User Manager portlet.
Roles assigned higher priority take precedence over lower-priority roles.
Alert Setup
Alert Setup
The Alert Setup portlet allows the Teradata Viewpoint Administrator to configure the Alert
Service to perform any of the following actions in response to alerts:
• Write to the alert log
• Send email notifications
• Send SNMP notifications
• Run BTEQ scripts
• Run programs
The Teradata Viewpoint Administrator can also configure the following alert presets:
• Core hours of operation
• Action sets of multiple alert actions run in a single operation
• Groups of multiple alert action sets run in a single operation
The Teradata Viewpoint Administrator can set time periods and file-size restrictions for retention of alert log data.
Settings configured in the Alert Setup portlet determine data displayed in the Alert Viewer and Teradata Systems portlets.
The Teradata Systems portlet allows you to add alerts to a system and set rules to trigger alerts based on the actions defined in the Alert Setup portlet.
Setting Alert Retention
You can set limits on how long to retain alert log data and how much to retain.
1
2
3
From the SETUP OPTIONS list, click Delivery Settings.
From the DELIVERY TYPES list, click Alert Log.
Under Delete Alert Log Data, do any of the following:
• Select the After check box to set the retention period using a number and calendar value.
• Select the Over check box, and then set the file-size restriction using the Over box and list.
32 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
4
Click Apply to apply the settings.
If you set a retention period, then after the specified period expires, alert log data is deleted in weekly increments. If you set a file-size restriction, then after the specified file size is exceeded, alert log data is deleted in weekly increments with the oldest data being deleted first.
Configuring Email Settings for Alerts
You can configure the Alert Service to send alert notifications by email.
1
2
3
4
5
6
7
8
9
10
From the SETUP OPTIONS list, click Delivery Settings.
From the DELIVERY TYPES list, click Email.
Under SMTP Host, do the following:
a
In the box, type the host address of the outgoing (SMTP) email server.
b
[Optional] Select the Use SSL check box to use Secure Sockets Layer as the transmission protocol.
Under Port, do one of the following:
• Select Use default port to use the default communications port on the SMTP host for outgoing email.
• Select Enter port number, and in the box, type the custom port number for the
SMTP host to use for outgoing email.
In the Server Timeout box, type the number of seconds the system must wait to establish a connection with the SMTP host before timing out.
The default is 30 seconds.
In the Reply-to box, type the email address to use as the Reply-to address in delivered alert email messages.
Under Login, do one of the following:
• Select Anonymous login to log on to the SMTP host anonymously.
• Select Enter credentials and in the boxes type a Username and Password to use to log on to the SMTP host.
Leave the Advanced box empty.
To test your email delivery settings, type a valid recipient email address in the Test
Recipient box and click Test.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
If the test is successful, an email message is sent to the test recipient address, addressed from the email address specified in the Reply-to box.
Click Apply to apply the settings.
Adding and Editing SNMP Configurations for Alerts
You can configure the Alert Service to deliver alert notifications to third-party management applications using SNMP.
Teradata Viewpoint User Guide, Release 13.11
33
Chapter 7 System Administration
1
2
3
4
5
6
From the SETUP OPTIONS list, click Delivery Settings.
From the DELIVERY TYPES list, click SNMP.
From the SNMP list, do one of the following:
• To add an SNMP configuration, click .
• To copy an SNMP configuration, click in the row of the configuration you want to copy.
• To edit an SNMP configuration, click the name of the configuration.
In the Configuration Name box, type a name for the SNMP configuration.
In the Destination box, type the IP address or host name of the destination for SNMP alert notifications.
In the Community box, type a name for the SNMP community.
The default is public
.
7
8
9
You also have the option to add or remove Destination and Community pairs.
To test the SNMP configuration on the specified destination host or hosts, click Test.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
If the test is successful, the Alert Service sends an SNMP trap (notification) to each specified destination. Verify that the SNMP trap is received at each destination.
[Optional] Select the Default for Alert Request Collector check box to make this the default configuration for SNMP requests originating from the AlertRequest and
MonitorRequest tables in the dbcmngr database.
Note: Making this configuration the default clears the previous default.
Click Apply to apply the settings.
Adding and Editing BTEQ Configurations for Alerts
You can configure the Alert Service to run BTEQ alert actions.
1
2
3
4
5
From the SETUP OPTIONS list, click Delivery Settings.
From the DELIVERY TYPES list, click BTEQ.
From the BTEQ list, do one of the following:
• To add a BTEQ configuration, click .
• To copy a BTEQ configuration, click in the row of the configuration you want to copy.
• To edit a BTEQ configuration, click the name of the configuration.
In the TDPID box, type the TDPID of the Teradata Database you want to log on to.
The BTEQ scripts are run against the specified TDPID. The TDPID must match the configuration in the Teradata Systems portlet.
[Optional] In the Account ID box, type the logon account ID under which to run the
BTEQ configuration.
34 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
6
7
8
9
10
The account ID is associated with the Username specified in this pane.
Select a Session Character Set from the list.
In the boxes, type a valid Username and Password to use to log on to the specified TDPID.
[Optional] Type a Logmech Name and Logmech Data in the boxes.
Logmech Name is the logon mechanism, and Logmech Data is the logon data
(parameters or credentials) for the selected mechanism.
To test the BTEQ configuration on the specified TDPID, click Test.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
Click Apply to apply the settings.
Deleting SNMP or BTEQ Configurations
You cannot delete a BTEQ configuration if it is used in an action set, or if it is the last configuration for a defined BTEQ alert action. You cannot delete an SNMP configuration if it is used in an action set.
1
2
3
4
From the SETUP OPTIONS list, click Delivery Settings.
From the DELIVERY TYPES list, click SNMP or BTEQ.
From the SNMP or BTEQ list, click to delete a configuration.
A confirmation message appears.
Click OK.
About Core Hours
The Alert Service operates 24 hours a day, 7 days a week.
You can schedule alert actions to run during one or more of the following times:
• Core hours
• Evening hours
• Weekend hours
Core hours are set as specific days of the week, with either a specific time interval or all day
(24 hours).
Evening hours and weekend hours are defined as follows:
•
Evening hours. On core days, the hours opposite the core hours.
• Weekend hours. The ending time of the ending core day to the beginning time of the beginning core day.
Time Interval
If the core hours are Monday through Friday, 9:00 AM to 5:00 PM, then evening hours are Monday through Friday, 5:00 PM to 9:00 AM, and weekend hours are from Friday at
5:00 PM through Monday at 9:00 AM.
Teradata Viewpoint User Guide, Release 13.11
35
Chapter 7 System Administration
All Day
If the core hours are Monday through Friday, 24 hours, then there are no evening hours, and weekend hours are Saturday through Sunday, 24 hours.
Setting Core Hours
As you make selections in this view, the EVENING HOURS and WEEKEND HOURS are updated in the view.
1
2
3
4
5
6
From the SETUP OPTIONS list, click Alert Presets.
From the PRESET OPTIONS list, click Core Hours.
In the CORE HOURS pane, under Days, select the beginning and ending core days from the two lists.
Under Time, do one of the following:
• Select 24 hours to specify the entire day on core days.
• Select the time interval option and select the beginning and ending times from the two lists.
From the Time Zone list, select the time zone to use for the core hours.
Click Apply to apply the settings.
Adding and Editing Action Sets
Alert actions can be combined into action sets. An action set allows you to run multiple alert actions in a single operation.
1
2
3
4
5
6
From the SETUP OPTIONS list, click Alert Presets.
From the PRESET OPTIONS list, click Action Sets.
From the ACTION SETS list, do one of the following:
• To add an action set, click .
• To copy an action set, click in the row of the action set you want to copy.
• To edit an action set, click the name of the action set.
In the Action Set Name box, type a name for the action set.
Under Times, select check boxes for the hours of operation during which you want the action set to run.
Under Actions, select any of the following actions by selecting the appropriate check box:
Option
Include in alert log
Description
Enables the alert log.
Send to person
In the appropriate box, type the recipient name or ID, role, or email address.
Specify multiple users in any box by separating them with a semicolon. You can use the boxes in any combination to specify different recipients.
36 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Option Description
SNMP
BTEQ
Select the SNMP configuration from the list.
From the lists, select the BTEQ script to run and the TDPID of the Teradata
Database to log on to and against which to run the script. Select Get from
alert to specify the Teradata Database that generated the alert.
Run a program From the list, select the program to run. You have the option to enter commandline arguments.
7
You also have the option to add or remove actions. The Send to person, SNMP,
BTEQ, and Run a program actions are dimmed if they have not been defined by the
Teradata Viewpoint Administrator or Teradata System Administrator. For additional information, see the Teradata Viewpoint Configuration Guide.
Click Apply to apply the settings.
Adding and Editing Groups
Alert actions can be combined into action sets, and alert action sets can be combined into groups. A group allows you to run multiple action sets in a single operation. Action sets must be defined before they can be included in a group.
1
2
3
4
5
6
From the SETUP OPTIONS list, click Alert Presets.
From the PRESET OPTIONS list, click Groups.
From the GROUPS list, do one of the following:
• To add a group, click .
• To copy a group, click in the row of the group you want to copy.
• To edit a group, click the name of the group.
In the Group Name box, type a name for the group.
From the list under Include the Following Action Sets, select the action set you want to include in the group.
You also have the option to add or remove action sets.
Click Apply to apply the settings.
Deleting Action Sets or Groups
If you delete a group, the group configuration is lost but the action sets combined in the group remain. You cannot delete an action set if it is used in a group or an alert rule. You cannot delete a group if it is used in an alert rule.
1
2
3
4
From the SETUP OPTIONS list, click Alert Presets.
From the PRESET OPTIONS list, click Action Sets or Groups.
From the ACTION SETS or GROUPS list, click to delete a configuration.
A confirmation message appears.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
37
Chapter 7 System Administration
Group Examples
Alert actions can be combined into action sets, and action sets can be combined into groups.
A group allows you to run multiple action sets in a single operation.
Different Alert Actions for Weekday and Weekend Hours
If you want a different set of alert actions to be performed for an event depending on the time period, define time-specific action sets and assign them to a group.
If you want the alert to perform an SNMP notification when a space-usage threshold is exceeded on a weekday, and to perform an SNMP notification and also send an email message to the Teradata Database Administrator when the threshold is exceeded during the weekend, define two action sets: one SNMP notification action for core, evening, and weekend hours, and one email notification action for weekend hours only.
Create a group and assign both action sets to the group. Then, in the Teradata Viewpoint administrative portlet where you want to define the alert, assign the group as the alert action.
Managing Email Alert Recipients
Groups and action sets can also be used to manage alert recipients, similar to email distribution lists.
Define action sets that send email notifications to different sets of people in your organization.
One action set could send an email notification to the Teradata Database Administrator, and another action set could send an email notification to end users in the finance department.
Create a group and assign both action sets to the group. Then, in the Teradata Viewpoint administrative portlet where you want to define the alert, assign the group as the alert action. When an alert is issued, both sets of users receive an email notification.
Backup
Backup
The Backup portlet allows you to enable automatic Teradata Viewpoint system backups and configure the data retention settings.
The Backup portlet displays the backup process icon with either a successful completion or an error for the last attempt and lists the last successful backup date and time. Use the
Backup portlet to:
• Enable and disable the daily backup process
• Store the backup files on the local system or at a network location
• Select the daily backup time
• Select how many days to keep the backup files
38 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
About Backup Errors
The BACKUP dialog box displays an out-of-service icon and the status message AN
ERROR HAS OCCURRED if an error occurs during the backup process. To investigate the error, you can view the backup log file:
/opt/teradata/dcs/logs/backup.log
. For additional information, see the Teradata Viewpoint Configuration Guide.
Scheduling Viewpoint Backups
You can configure automatic backups, set a data retention length, and enable or disable the backup process at any time.
1
2
3
4
5
[Optional] From the BACKUP view, do one of the following:
• Select the Enable backup check box to activate the backup process.
• Clear the Enable backup check box to disable the automatic-backup process.
Under Location, select one of the following options to set the location of the backup files:
• Click Local to use the local server.
• Click Network to use a network location and do the following:
• Enter a Host to specify the host system name.
• Enter a Path to specify the absolute path name where the backups are stored on the host system.
• [Optional] Click Test to verify that the login settings are correct. The icon appears if the operation is successful. The icon appears if the operation fails.
Verify that the settings are correct, and try again.
From the Daily Backup Time list, select the daily start time of the backup operation.
From the Keep Backups For list, select the number of days to retain the backup data.
Click Apply.
The changes are saved and backup files are deleted when they are older than the configured number of days.
LDAP Servers
LDAP Servers
The LDAP Servers portlet allows you to configure LDAP servers to authenticate users and assign user roles in Teradata Viewpoint. The LDAP Servers portlet displays the names of currently configured LDAP servers or is empty if an LDAP is not configured.
• Use next to SETUPS to add an LDAP server.
• Use the order buttons to adjust the evaluation order of the LDAP-server list.
• Use next to the server name to delete an LDAP server.
Teradata Viewpoint User Guide, Release 13.11
39
Chapter 7 System Administration
• Use the dialog box to enable and disable an LDAP server, and configure the LDAP to automatically validate users against the LDAP directory with the auto-provisioning feature or assign user roles with the role mapping feature.
What is an LDAP Server?
Teradata Viewpoint uses a Lightweight Directory Access Protocol (LDAP) server to:
• Validate new user information by querying the LDAP for authentication. The binding operation authenticates a username and a password.
• Automatically create a Teradata Viewpoint username with the auto-provisioning feature.
• Automatically assign users to a Teradata Viewpoint role with the Role Mapping feature.
Adding an LDAP Server
You can add an LDAP server to authenticate and assign roles to Teradata Viewpoint users.
1
2
3
4
5
6
7
Click next to SETUPS.
Enter a Nickname for the LDAP server (up to 8 characters).
[Optional] Select the Enabled check box to activate LDAP upon successfully adding the
LDAP server to Teradata Viewpoint.
Enter a URL address.
One URL address is required. Use the evaluation order.
to delete, add, and move up or move down in
Select the Name Matching method from the following options:
• Select DN Pattern Bind to specify one or more patterns that represent a distinguished name (DN) in the LDAP directory.
• Select User Search to search for a pattern in the base directory or subdirectories:
• [Optional] Enter a Service account DN and Service account password to search with a specific account.
• [Optional] Leave Service account DN and Service account password blank to connect to the LDAP as an anonymous user.
• Enter a Search pattern.
• Enter a Search base to specify a base directory for the search pattern.
• [Optional] Select the Search extent Recursive scan check box to include subdirectories of the base directory in the search.
• [Optional] Under Search extent, clear the Recursive scan check box to limit the search to the specified base directory.
Configure Key User Information:
• Enter the LDAP first name attribute.
• Enter the LDAP last name attribute.
• Enter the LDAP email attribute.
Every user has an email address. This default email domain is used if a valid email address is not found in the LDAP directory during auto-provisioning.
[Optional] Test your settings:
40 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
8
9
10 a b
Click Test to display the LDAP test panel.
Enter a Test username and Test password.
c
Click Run.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
[Optional] Configure Auto-Provisioning:
a b
[Optional] Select the Turn on auto-provisioning check box to automatically validate users against the LDAP directory.
Select a role from the Automatically assign these roles menu. You can add or remove roles.
c
Enter a Default email domain to provide a notification email address.
Every user has an email address. This default email domain is used if a valid email address is not found in the LDAP directory during auto-provisioning.
If the group role mapping method is specified, configure Role Mapping:
a
Enter a Group search base.
b c
[Optional] Select the Group search subtree check box to include subtree directories during the binding process.
Enter a Group attribute name.
d
[Optional] Click Add Role Mapping to add attribute or group settings so that you can assign another role.
Click Apply.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
About Role Mapping
In Teradata Viewpoint, users are assigned roles to organize and control their access to portlets, metrics, and features. You can use the Roles Manager portlet to assign permissions to the roles. During the process of adding an LDAP configuration to Teradata Viewpoint, the optional Role Mapping section in the LDAP Servers portlet allows you to assign a role automatically to auto-provisioned users.
Use role mapping to:
• Assign roles for a new user.
• Add roles that were assigned to the user since the last user login.
• Remove roles that were removed since the last user login.
Consider the following:
• Only roles already defined using the Roles Manager portlet can be mapped.
Teradata Viewpoint User Guide, Release 13.11
41
Chapter 7 System Administration
• The Teradata Viewpoint Administrator can change a role at any time using the Roles
Manager portlet.
Role Mapping Use Cases
In the following use cases, the LDAP value represents the distinguished name (DN) of a group. When the Type is Attribute, the attribute name field represents the user attribute where the group DN is stored. When the Type is Group, the group attribute name field is used for mapping and represents the group attribute where the user DN is stored.
The first use case assigns users to a Teradata Viewpoint role based on an attribute of the user's record in LDAP. Assume you have a user with the following record in LDAP: dn: cn=User1,ou=Users,dc=teradata,dc=com memberOf: cn=USA,ou=Groups,dc=teradata,dc=com
In this case, the Group attribute name field is not relevant. To map User1 to the Teradata
Viewpoint role US Users, create the following role mapping:
• Type = Attribute
• Attribute name = memberOf
•
LDAP value = cn=USA,ou=Groups,dc=teradata,dc=com
• Viewpoint role = US Users
The second use case assigns users to a Teradata Viewpoint role based on membership in an LDAP group with the following structure: dn: cn=DBAs,ou=Groups,dc=teradata,dc=com uniqueMember: cn=User2,ou=Users,dc=teradata,dc=com
In this case, set the Group attribute name field to uniqueMember. To map User2 to the
Teradata Viewpoint role DBAs, create the following role mapping:
• Type = Group
•
LDAP value = cn=DBAs,ou=Groups,dc=teradata,dc=com
• Viewpoint role = DBAs
Managed Servers
Managed Servers
A managed server is a server in a Teradata cabinet that is monitored by the same Teradata server management solution used to monitor TPA nodes. A managed server might be used for storage, backup, or housing data collectors that provide portlet information and metric data. Teradata Viewpoint monitors managed servers for system-level metrics and collects information for Teradata Viewpoint portlets.
The Managed Servers portlet displays the names of currently configured managed servers or is empty if no managed server is configured. Use next to SERVERS to add a managed server. Use the dialog box to delete, enable, and disable a managed server for use by the
Teradata Viewpoint Data Collection Service (DCS).
42 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Adding a Managed Server
You can add a managed server to Teradata Viewpoint. After a managed server is defined in
Teradata Viewpoint, you can test the configuration settings and enable the managed server for use by the Data Collection Service (DCS).
1
2
3
4
Click next to SERVERS.
[Optional] Select the Server enabled check box to activate the managed server for the DCS.
Enter values for these required fields:
• Managed Server Name is an 8-character alphanumeric name.
•
Hostname is the managed server IP address or domain name.
• Login is the user name and password.
The default values only need to be changed if you change the login and password settings during the installation of TMS Monitor.
Do any of the following:
• Click Test to verify that the login settings are correct.
• Click Apply to update the managed server.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again. If you are unable to enable or disable the managed server, contact the Teradata Viewpoint Administrator.
Deleting a Managed Server
You can delete a Teradata Viewpoint managed server. This is a permanent procedure that differs from disabling the managed server from the Data Collection Service.
1
2
3
From the SERVERS list, click on the managed server name you want to remove.
In the dialog box, click Delete Server.
A confirmation message appears.
Click OK.
Enabling or Disabling a Managed Server
After a managed server is defined in Teradata Viewpoint, you can enable it for use by the
Data Collection Service (DCS) or disable it.
1
2
3
From the SERVERS list, click on the managed server name you want to update.
Do one of the following:
• Select the Server enabled check box to activate a managed server.
• Clear the Server enabled check box to disable a managed server.
When disabled, the DCS does not use the managed server. Any existing data about the managed server can still be viewed.
Do any of the following:
Teradata Viewpoint User Guide, Release 13.11
43
Chapter 7 System Administration
• Click Test to verify that the login settings are correct.
• Click Apply to update the managed server.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again. If you are unable to enable or disable the managed server, contact the Teradata Viewpoint Administrator.
Portlet Library
Portlet Library
The Portlet Library allows you to enable or disable portlets globally. This setting takes precedence over every other portlet permission level.
The Manage Existing Portlets view allows you to manage the installed portlets, which are grouped by category. Details about portlets are also provided:
• Parent portlet name
• Version number
• Publisher name
• Bundle name
• Installation date
• Whether the portlet is enabled
• Whether the portlet can be deleted
Portlets in the Admin category display N/A in the ENABLE column. These portlets cannot be disabled and are included in the view only to provide version information.
A shared portlet is a user-defined version of a portlet. The Parent Portlet column identifies the original portlet before its customization as a shared portlet. Only portlets in the Shared
Portlet category can be deleted.
Enabling or Disabling Portlets
The Teradata Viewpoint Administrator can use the Portlet Library to enable or disable installed portlets for Teradata Viewpoint user access.
1
2
3
In the Manage Existing Portlets dialog box, click a name to expand a category.
For each installed portlet, do one of the following:
• Enable the portlet by selecting the corresponding check box.
• Disable the portlet by clearing the corresponding check box.
You can enable all the portlets by clicking the ENABLE column heading. Portlets in the
Admin category display N/A in the ENABLE column. These portlets cannot be disabled and are included in the view only to provide version information.
Click Apply.
44 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Deleting a Shared Portlet
The Teradata Viewpoint Administrator can use the Portlet Library to delete a shared portlet so Teradata Viewpoint users can no longer access it.
1
2
3
In the Manage Existing Portlets dialog box, click Shared to expand the category.
Click under the DELETE column that corresponds to the shared portlet you want to delete.
A confirmation message appears.
Click OK.
The portlet name no longer appears in the Add Content menu in the portal, and the portlet is no longer available to Teradata Viewpoint users. Existing instances of a shared portlet on a portal page remain on the page. However, a warning message appears if a user tries to access the portlet.
Roles Manager
Roles Manager
The Roles Manager portlet allows the Teradata Viewpoint Administrator to assign permissions efficiently by creating classes of users called roles.
The Teradata Viewpoint Administrator can perform the following tasks:
• Add and configure new roles
• Edit the configuration and settings of existing and default roles
• Copy roles, saving time in creating new roles
• Enable or disable portlets for a role.
• Delete roles that are no longer needed
About Roles
Teradata Viewpoint includes preconfigured roles that cannot be removed from Teradata
Viewpoint. Change the following roles according to your needs:
Administrator
This role has all permissions and can be assigned to any account. It is recommended that this role be used only by the Teradata Viewpoint Administrator.
User
This role is assigned to every Teradata Viewpoint user. It is recommended that this role be set with no permissions.
Teradata Viewpoint User Guide, Release 13.11
45
Chapter 7 System Administration
TD_SE (Teradata Solution Engineer)
This role is intended for the Teradata Solution Engineer when they need access to system health and metrics. It is recommended that you enable or disable this role, but do not reconfigure it.
TD_TSS (Teradata Technical Support Specialist)
This role is intended for the Teradata Technical Support Specialist when they need access to your system SQL. It is recommended that you enable or disable this role, but do not reconfigure it.
It is recommended that you copy Administrator and User to configure new roles with partial permissions that are appropriate to all users in that role. Each role you create controls access to specific systems, portlets, metrics, preferences, and permissions in portlets.
Example Roles
The following examples show how roles can be defined to manage portlet usage and control permissions within Teradata Viewpoint.
AppDBA
The AppDBA role is defined for users that are application DBAs. The rewind feature and some portlets are enabled for this role so users can review graphs in the Productivity portlet.
AcctDept
The AcctDept role is defined for all users in the Accounting Department. Some portlets are enabled for this role. These users are granted permissions so they can edit all calendar events in the Calendar portlet and access the My Queries portlet to monitor their queries.
OpDBA
The OpDBA role is defined for users that are operations DBAs. The rewind feature and all portlets are enabled for this role so users can monitor performance trends. Users in the
OpDBA role access the Workload Monitor portlet to track Teradata Database request arrivals and completions.
About the Roles Manager View
The Roles Manager view allows the Teradata Viewpoint Administrator to manage roles efficiently, including assigning users to roles, and granting permissions to roles. After a role is created, you can customize the role using the following tabs:
General
Enable or disable a role, and select the systems available to that role.
Users
Search for users and add them to, or remove them from, the list of users assigned to the selected role.
46 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Portlets
Enable or disable portlets for a role. This tab can also be used to configure default settings per portlet, per role.
Metrics
Select the metrics to display per metric, per system.
Permissions
Control whether users in this role can set their own preferences and share customized versions of the portlet with other users, and configure which portlet features the role can access.
Note: The SET PREFS check box is only available for portlets that have a
PREFERENCES view.
Adding a Role
You can add a role and assign one or more Teradata Database systems to the role.
1
2
3
4
5
6
From the Roles Manager view, click Add Role.
An Add role dialog box appears with the General tab selected (default).
Under Name, enter a new name for the role, up to 25 characters.
You can use alphanumeric characters and underscore (_), but no space.
Under Description, use the default or enter a new description, up to 255 characters.
[Optional] Under Role State, do one of the following:
• Leave the Role State as is.
• Select Enabled to enable the role for configuration and adding users, and activate the role for use with Teradata Viewpoint.
• Select Disabled to enable the role for configuration and adding users, but disable the role for use with Teradata Viewpoint.
[Optional] Under Rewind State, do one of the following:
• Leave the Rewind State as is.
• Select Enabled to grant permission to this role to use the rewind feature, when available.
• Select Disabled to deny permission for this role to use the rewind feature.
[Optional] Under Enable Systems For Role, do one of the following:
• Leave the currently selected Teradata Database systems as is.
• Select new Teradata Database systems the selected role is authorized to access.
• Click the Enable Systems For Role column heading to select all check boxes for the
Teradata Database systems.
Note: You can click the column heading again to clear all check boxes.
Editing a Role
You can edit a role, but you cannot change the name of an existing role. To change the name, you must either add a new role or copy the role, and then change the role name.
Teradata Viewpoint User Guide, Release 13.11
47
Chapter 7 System Administration
1
2
3
4
5
From the Roles Manager view, select a role from the list.
A dialog box appears with the current information for the selected role.
Under Description, use the default or enter a new description, up to 255 characters.
[Optional] Under Role State, do one of the following:
• Leave the Role State as is.
• Select Enabled to enable the role for configuration and adding users, and activate the role for use with Teradata Viewpoint.
• Select Disabled to enable the role for configuration and adding users, but disable the role for use with Teradata Viewpoint.
[Optional] Under Rewind State, do one of the following:
• Leave the Rewind State as is.
• Select Enabled to grant permission to this role to use the rewind feature, when available.
• Select Disabled to deny permission for this role to use the rewind feature.
[Optional] Under Enable Systems For Role, do one of the following:
• Leave the currently selected Teradata Database systems as is.
• Select new Teradata Database systems the selected role is authorized to access.
• Click the Enable Systems For Role column heading to select all check boxes for the
Teradata Database systems.
Note: You can click the column heading again to clear all check boxes.
Copying a Role
When a role is copied, all settings on all tabs are inherited from the original role, except for the Name field on the General tab. When the Administrator role is copied, all information on all tabs is copied. However, the copied role does not have administrative privileges and does not have an Admin menu on the portal.
1
2
3
4
5
6
From the Roles Manager view, select a role from the list.
Click Copy Role.
A dialog box appears with the General tab selected.
Under Name, enter a new name for the role, up to 25 characters.
You can use alphanumeric characters and underscore (_), but no space.
Under Description, use the default or enter a new description, up to 255 characters.
[Optional] Under Role State, do one of the following:
• Leave the Role State as is.
• Select Enabled to enable the role for configuration and adding users, and activate the role for use with Teradata Viewpoint.
• Select Disabled to enable the role for configuration and adding users, but disable the role for use with Teradata Viewpoint.
[Optional] Under Rewind State, do one of the following:
48 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
7
• Leave the Rewind State as is.
• Select Enabled to grant permission to this role to use the rewind feature, when available.
• Select Disabled to deny permission for this role to use the rewind feature.
[Optional] Under Enable Systems For Role, do one of the following:
• Leave the currently selected Teradata Database systems as is.
• Select new Teradata Database systems the selected role is authorized to access.
• Click the Enable Systems For Role column heading to select all check boxes for the
Teradata Database systems.
Note: You can click the column heading again to clear all check boxes.
Deleting a Role
You can delete a role. After a role is deleted, it cannot be restored. This action affects all users assigned to the role.
1
2
3
From the Roles Manager view, select a role from the list.
Click Delete Role.
A confirmation message appears.
Click OK.
The deleted role is removed from the list.
Assigning Users to a Role
You can search for users and assign them to a role.
If auto-provisioning is in use, a default role is automatically assigned to new users the first time they log on. User is generally the default assignment.
1
2
3
4
5
From the Roles Manager view, select a role from the list.
Click the Users tab.
The users list appears and includes two panes. The left pane lists all Teradata Viewpoint system users. The right pane lists users already assigned to the selected role. In the left pane, the names of users already assigned to the selected role are dimmed.
[Optional] Search for a user:
a
Select a filter from the list.
b
In the contains field, enter the search criteria.
c
Click Find.
The AVAILABLE USERS list displays only users meeting the search criteria.
Select one or more users from the left pane.
To select multiple users, press Ctrl while selecting additional names.
Click .
Teradata Viewpoint User Guide, Release 13.11
49
Chapter 7 System Administration
50
6
The selected users are assigned to the role. Selected user names are copied to the right pane and are dimmed in the left pane.
Click Apply.
Removing Users from a Role
You can remove users from a role.
1
2
3
4
5
From the Roles Manager view, select a role from the list.
Click the Users tab.
The users list appears and includes two panes. The left pane lists all Teradata Viewpoint system users. The right pane lists users already assigned to the selected role. In the left pane, the names of users already assigned to the selected role are dimmed.
Select one or more users from the right pane.
You can select multiple users by holding the CRTL key and clicking additional names.
Click .
The selected users are removed from the role. The names no longer appear in the right pane and are no longer dimmed in the left pane.
Click Apply.
Portlet Preferences and Permissions
The Teradata Viewpoint Administrator can set permissions for a role to access portlets, preferences, and tabs. Higher level permissions override lower level permissions.
Portlet Access
This option is the highest permission level for portlet access and gives this role access to the entire portlet. To grant a role access to this portlet from the Add Content menu, select the ENABLE FOR ROLE check box from the Portlets tab.
Preferences Access
This option is the highest permission level for preferences and gives this role access to change all preferences in this portlet. To grant a role access to the
PREFERENCES view, select the SET PREFS check box from the Permissions tab.
Note: The SET PREFS check box is only available for portlets that have a
PREFERENCES view.
Tab Access
This option is the lowest permission level for preferences and denies this role access to tabs in the PREFERENCES view. To deny a role from using any tab in the
PREFERENCES view, select the Disable User Preferences for tab name globally check box from the Portlets tab by clicking under CONFIGURE DEFAULTS.
Enabling Portlets for a Role
You can enable Teradata Viewpoint portlets for a role. Portlets enabled using the Roles
Manager portlet are the default portlets available for all users assigned to the selected role.
Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
1
2
3
4
5
From the Roles Manager view, select a role from the list.
Click the Portlets tab.
A dialog box appears displaying the portlet categories.
Click a name to expand a category and display the available portlets.
For the selected role, select the portlets to enable in each category.
Click the ENABLE FOR ROLE column heading to enable the role for all portlets. Click the column heading again to clear all check boxes. Enabling a portlet in the Roles
Manager view, Portlets tab, includes the portlet in the Teradata Viewpoint Add
Content menu for all role-assigned users.
Click Apply.
Configuring Default Portlet Settings
You can configure default portlet settings for a role.
Shared portlets are preconfigured using an instance of a portlet and cannot be configured using the Roles Manager portlet.
1
2
3
4
From the Roles Manager view, select a role from the list.
Click the Portlets tab.
A dialog box appears displaying the portlet categories.
Click a name to expand a category and display the available portlets.
Configure defaults for each portlet as follows:
5 a
Click for the portlet you want to configure.
The CONFIG view for the corresponding portlet appears.
Configuration defaults entered in the CONFIG view only apply to the role being configured. If a user is assigned to multiple roles, the role with the highest priority determines which settings are applied to every instance of the portlet added to a portal page. For example, if a user is assigned the roles of Administrator then User, the Administrator settings apply to each instance of the portlet. Also, the Teradata
Viewpoint Administrator can grant users permission to set preferences for a subset of the available settings as long as user preferences are not disabled.
b
Follow the instructions in the preferences-related topics provided for each portlet in
Teradata Viewpoint Help.
Click Apply.
Granting Metric Permissions for a Role
You can grant permissions for metrics by role. The metric permissions granted to a role apply to the following portlets only:
•
Capacity Heatmap
•
Metrics Analysis
•
Metrics Graph
Teradata Viewpoint User Guide, Release 13.11
51
Chapter 7 System Administration
•
Todays Statistics
1
2
3
4
From the Roles Manager view, select a role from the list.
Click the Metrics tab.
A dialog box appears displaying the metrics categories and each Teradata Database system name.
Select the metrics for each available Teradata Database system that you want users in the selected role to see by doing any of the following:
• Click the top check box in a column to select or clear all check boxes for a category for all systems.
• Click the system name at the left of the row to select or clear all check boxes for that system.
• Click a check box to select or clear that category for that system.
Click Apply.
Granting Portlet Permissions for a Role
You can enable specific features of a portlet by system and role.
1
2
From the Roles Manager view, select a role from the list.
Click the Permissions tab.
3
4
5
Select a portlet from the Select portlet list.
The Permissions dialog box displays a list of the available systems and options.
Select the permissions you want to grant to the selected role for each available Teradata
Database system by doing any of the following:
• Select the SET PREFS check box to allow users to set their own preferences.
• Select the SHARE PORTLET check box to allow users to share customized versions of the portlet with other users.
• Click the top check box in a column to select or clear all check boxes for a category for all systems.
• Click the system name at the left of the row to select or clear all check boxes for that system.
• Click a check box to select or clear that category for that system.
Click Apply.
Portlet Permissions
The Teradata Viewpoint Administrator can grant access to portlet views using the
Permissions tab. You must select the parent view setting for dependent child views to display.
52 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Permission Name
SET PREFS
SHARE PORTLET
ABORT HOST
ABORT QUERY
ACCOUNT
ACCOUNT
DRILLDOWN
Description
Enables role access to set their own preferences in this portlet.
After this setting is enabled, the user can click in the portlet.
Enables role to share customized versions of the portlet with other users. After this setting is enabled, the user can click in the portlet.
Enables role access to the Abort
Host console utility.
Enables role to abort queries from the Tools menu.
Enables role to select By Account
String in the SELECT REPORT dialog box.
Enables role access to a table of all sessions for each account.
Dependency
• VIEW
SUMMARY
• VIEW
SUMMARY
• ACCOUNT
ADD SPACE Enables role to reallocate permanent disk space from one database to another.
ALL SESSIONS REPORT Enables role to select By Session >
All in the SELECT REPORT dialog box.
ALLOCATION GRID Enables role access to the
Cylinder Allocation by
Temperature and Grade view.
CHANGE PRIORITY Enables role to change the priority of a query or session from the
Tools menu.
CHANGE STATE Enables role to change a state.
CHANGE WORKLOAD Enables role to change the workload of a query or session from the Tools menu.
CHECKTABLE Enables role access to the Check
Table console utility.
CLEAR ALERT Enables role to clear an alert.
CONFIGURE Enables role access to the
Configure console utility.
Portlets Displaying Setting
The SET PREFS check box is only available for portlets that have a
PREFERENCES view.
The SHARE PORTLET check box is only available for portlets that can be shared.
Remote Console
My Queries, Query Monitor, Workload
Monitor
Query Monitor
Query Monitor
Space Usage
Query Monitor
TVS Monitor
My Queries, Query Monitor, Workload
Monitor
Application Health, Ecosystem Health,
Process Health, Table Health
My Queries, Query Monitor, Workload
Monitor
Remote Console
Application Health, Ecosystem Health,
Log Viewer, Process Health, Table Health
Remote Console
Teradata Viewpoint User Guide, Release 13.11
53
Chapter 7 System Administration
Permission Name
CREATE EVENT
CURRENT
DISTRIBUTION
DBS CONTROL
EDIT ANY EVENT
Description
Enables role to create calendar events.
Enables role access to the
Cylinder Distribution by Grade
view.
Enables role access to the DBS
Control console utility.
Enables role to edit any events, including events created by other users.
EDIT OWN EVENT
EDIT RULESETS
Enables role to only edit their own calendar events.
Enables role to edit rulesets
FERRET Enables role access to the Ferret console utility.
GATEWAY GLOBAL Enables role access to the Gateway
Global console utility.
HISTORICAL
DISTRIBUTION
LOCK DISPLAY
Enables role access to the
Historical Cylinder Distribution
by Grade view.
Enables role access to the Lock
Display console utility.
MY CRITERIA Enables role to select By Session >
My Criteria in the SELECT
REPORT dialog box.
OPERATOR CONSOLE Enables role access to the
Operator Console console utility.
PARTITION
DRILLDOWN
PRIORITY
SCHEDULER
QUERY
CONFIGURATION
QUERY SESSION
Enables role to select a utility from the NAME column in the utility view to see all sessions for the selected utility.
Enables role access to the Priority
Scheduler console utility.
Enables role access to the Query
Configuration console utility.
RECOVERY
MANAGER
RELEASE QUERY
Enables role access to the Query
Session console utility.
Enables role access to the
Recovery Manager console utility.
Enables role to release a query that is queued and waiting to run from the Tools menu.
Dependency
• VIEW
SUMMARY
• UTILITY
Portlets Displaying Setting
Calendar
TVS Monitor
Remote Console
Calendar
Calendar
Workload Designer
Remote Console
Remote Console
TVS Monitor
Remote Console
Query Monitor
Remote Console
Query Monitor
Remote Console
Remote Console
Remote Console
Remote Console
My Queries, Query Monitor, Workload
Monitor
54 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Permission Name Description
SESSION DRILLDOWN Enables role to view details of each session from the Account String,
Session, User, or Utility reports.
Dependency
• VIEW
SUMMARY
• ACCOUNT
• ACCOUNT
DRILLDOWN
• ALL
• MY CRITERIA
• USER
• USER
DRILLDOWN
• UTILITY
• PARTITION
DRILLDOWN
Portlets Displaying Setting
Query Monitor
Remote Console SHOW LOCKS
SPACE BY DATABASE Enables role to select Database:
By most space in the SELECT
REPORT dialog box.
SPACE BY TABLE Enables role access to the details view with a list of tables in each database.
SPACE BY VPROC
Enables role access to the Show
Locks console utility.
Enables role to select Vproc: By
space in the SELECT REPORT dialog box.
TERADATA DWM
USER
Enables role access to the Teradata
DWM Dump console utility.
Enables role to select By User in the SELECT REPORT dialog box.
• VIEW
SUMMARY
Space Usage
Space Usage
Space Usage
Remote Console
Query Monitor
Query Monitor USER DRILLDOWN Enables role access to session details for that user.
• VIEW
SUMMARY
• USER
UTILITY
VIEW ACTIONABLE
MENU
Enables role access to the Tools menu in the summary view.
VIEW ALERT DETAIL Enables role access to the details view.
VIEW BLOCK
Enables role to select By Utility in the SELECT REPORT dialog box.
Enables role to see that a query is blocked. The Blocked By tab is visible, but contains no details.
• VIEW
SUMMARY
Query Monitor
My Queries, Query Monitor, Workload
Monitor
Alert Viewer
SQL Scratchpad
Teradata Viewpoint User Guide, Release 13.11
55
Chapter 7 System Administration
Permission Name Description
VIEW BLOCK DETAILS Enables role access to the content of the Blocked By tab.
Dependency
• VIEW
SUMMARY
• VIEW BLOCK
Portlets Displaying Setting
SQL Scratchpad
VIEW BLOCKED BY Enables role access to the Blocked
By tab.
• VIEW
SUMMARY
• SESSION
DRILLDOWN
My Queries, Query Monitor, Query
Spotlight, Workload Monitor
VIEW CANARY
QUERY
Enables role to view canary queries in the summary view.
VIEW DBQL METRICS Enables role to view the Queries
Per Hour metrics.
VIEW DELAY
Enables role access to the Delay tab. • VIEW
SUMMARY
• SESSION
DRILLDOWN
Canary Response Times, Productivity
Productivity
My Queries, Query Monitor, Workload
Monitor
VIEW DETAIL Enables role access to the details view.
VIEW EXPLAIN
VIEW EXPLAIN
SIMPLE
Enables role access to the Explain tab.
Enables role access to the Explain tab.
• VIEW
SUMMARY
• VIEW
SUMMARY
• SESSION
DRILLDOWN
My Queries, Query Spotlight, System
Health, Workload Health, Workload
Monitor
My Queries, Query Monitor, Query
Spotlight, Workload Monitor
• VIEW
SUMMARY
SQL Scratchpad
VIEW NODES
VIEW NODE DETAIL Enables role access to the details
VIEW QUERY BAND
VIEW RULESETS
VIEW SQL
Enables role access to the summary view to see nodes.
view to see nodes.
Enables role access to the Query
Band tab.
Node Resources
• VIEW
SUMMARY
• VIEW NODES
Node Resources
• VIEW
SUMMARY
• SESSION
DRILLDOWN
My Queries, Query Monitor, Query
Spotlight, Workload Monitor
Enables role to view rulesets.
Enables role access to the SQL tab. • VIEW
SUMMARY
• SESSION
DRILLDOWN
Workload Designer
My Queries, Query Monitor, Query
Spotlight, Workload Monitor
56 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Permission Name
VIEW STATISTICS
Description
Enables role access to see statistics in the TODAYS STATISTICS view.
Enables role access to the summary view.
Dependency Portlets Displaying Setting
Todays Statistics
VIEW SUMMARY
VIEW VPROCS Enables role access to the summary view to see vprocs.
VIEW VPROC DETAIL Enables role access to the details view to see vprocs.
VPROC MANAGER Enables role access to the Vproc
Manager console utility.
• VIEW
SUMMARY
• VIEW
VPROCS
Canary Response Times, Lock Viewer,
My Queries, Node Resources,
Productivity, Query Monitor, Query
Spotlight, Remote Console, Space Usage,
SQL Scratchpad, System Health, TVS
Monitor, Viewpoint Monitoring,
Workload Health, Workload Monitor
Node Resources
Node Resources
Remote Console
Teradata Systems
Teradata Systems
The Teradata Systems portlet allows the Teradata Viewpoint Administrator to add, configure, enable, and disable Teradata Database systems using specific dialog boxes:
General
Configure the system nickname, TDPID, login names, passwords (hidden), and account strings (optional). Test the connection to Teradata Database, and add or delete login names.
Data Collectors
Enable, disable, and configure data collectors to capture and retain portlet, disk usage, and resource data.
System Health
Enable metrics for the System Health portlet. Configure degraded and critical thresholds for each metric.
Canary Queries
Configure canary queries used to test Teradata Database system response times. The
SYSTEM HEARTBEAT canary query cannot be removed.
Teradata Viewpoint User Guide, Release 13.11
57
Chapter 7 System Administration
Alerts
Add, delete, copy, and configure alerts, or migrate existing Teradata Manager alerts.
Monitor Rates
Set Teradata Database internal sample rates for sessions, node logging, and vproc logging.
Log Table Clean Up
Select system log tables to clean up.
Clean Up Schedule
Schedule clean up of system log tables.
You can view the amount of disk space used:
Disk usage
View the percentage of total disk space being consumed by collected data on the managed server.
Twelve hour change
View the percentage of disk space used or has become available over the past twelve hours. Positive indicates the amount of space used has increased. Negative indicates the amount of space used has decreased.
Adding a Teradata Database System
You can add a Teradata Database System to Teradata Viewpoint. The system must be added before you can configure the data collectors to monitor the database.
1
2
3
4
5
6
7
Click next to SYSTEMS.
Enter a SYSTEM NICKNAME (up to 8 characters).
[Optional] Select the System Enabled check box to activate the Teradata Database system for monitoring.
Enter the TDPID of the Teradata Database system.
Enter a LOGIN name and password.
You also have the option to specify an account string, add more logins, and test that the login settings are correct.
[Optional] Select the Enable data collectors check box to activate all data collectors except Virtual Storage so they can collect data points on this system.
The data collectors can be enabled and configured individually in Data Collectors. The
Virtual Storage data collector can only be enabled and configured in Data Collectors.
[Optional] In the CHARACTER SET section, select default character sets and enter a
JDBC Flag value.
• [Optional] Select a character set from the Session menu.
• [Optional] Select a character set from the Monitor menu.
58 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
8
• [Optional] Enter a JDBC Flag value.
Click Apply to apply the settings.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
Deleting a Teradata Database System
Deleting a Teradata Database system removes the system from Teradata Viewpoint and removes all data collector and canary query information.
1
2
3
4
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click General.
Click Delete System.
A confirmation message appears.
Click OK.
Adding a Login Account
After a Teradata System has been configured in Teradata Viewpoint, an unlimited number of login accounts can be added and referenced on other Teradata Systems portlet views.
The tdwm login is required to load and activate rulesets from Teradata Database. The tdwm login cannot be removed or renamed, but you can enter a password, enter an account string, and test the tdwm login settings.
1
2
3
4
5
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click General.
Under LOGIN, click .
A blank row appears.
Enter a LOGIN name and password.
You also have the option to specify an account string, add more logins, and test that the login settings are correct.
Click Apply to apply the settings.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
Deleting a Login Account
At least one login account must be configured for a Teradata Database system. You cannot remove the last login account or the tdwm login.
1
2
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click General.
3
Under LOGIN, click on the login row to be removed.
The row disappears. The login is removed permanently.
Teradata Viewpoint User Guide, Release 13.11
59
Chapter 7 System Administration
CHARACTER
SET Fields
Session
Standard Character Set Configurations
When you configure a Teradata Database system, the CHARACTER SET section of the
GENERAL setup allows you to define the charset that the system uses to communicate with
Teradata Viewpoint. The character set definitions are optional. If you do not define them,
Teradata Viewpoint uses the Teradata Database system settings.
Character Set Definition Suggestions
Monitor
JDBC Flag
The UTF8 session character set is strongly recommended for the Teradata Viewpoint session to ensure end-to-end fidelity of character data by avoiding conversion between character sets. However, the UTF8 session character set does not support the KANJI1 server character sets, which are used for storing data dictionary data in Teradata V2R6.x Japanese systems. For these systems, Japanese users must specify the
Kanji session character set.
The UNICODE session character set is not currently supported for the Teradata Performance Monitor and Console Partition. To create Teradata Viewpoint object names that connect Monitor and Console
Partitions, use Non-Unicode session character sets such as LATIN1252_0A, KANJISJIS_OS, and
KANJIEUC_OU.
The JDBC Flag setting is a legacy-support feature to assist the transition away from the unsupported use of the Teradata Database to store non-Latin characters in a Latin column, and the unsupported access of non-ASCII in an ASCII session character set. The JDBC Flag option specifies the CLIENT_CHARSET connection parameter, which can be used to override the Teradata JDBC Driver fixed-mapping of the
Teradata session character set to the Java character set. The valid values are the actual Java character set names. Here are two examples of valid JDBC Flag settings:
• Windows-936 from Chinese (PRC)
• Windows-950 for Chinese (Taiwan)
Teradata
Database
Version
V12.00 and later
V12.00 and later
V2R6.x
Language Support
Mode
Standard
Japanese
Standard
The following table lists the default Teradata Database system settings.
Data Dictionary
Charset
UNICODE
UNICODE
LATIN
Teradata
Viewpoint Session
Charset
UTF8
UTF8
Teradata
Viewpoint
Monitor Charset Limitations
ASCII or
LATIN1252_0A
ASCII session only supports the limited 7-bit
ASCII characters
KANJISJIS_OS Unknown
UTF8 Unknown
V2R6.x
Japanese KANJI1 KANJISJIS_OS or
KANJIEUC_OU
ASCII or
LATIN1252_0A
KANJISJIS_OS or
KANJIEUC_OU
Use either KANJISJIS_OS or KANJIEUC_OU, for both settings. Do not mix character sets.
Configuring Data Collectors
After a Teradata System has been configured in Teradata Viewpoint, data collectors can be configured to monitor the database.
60 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
1
2
3
4
5
6
7
8
9
10
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Data Collectors.
Click the data collector name you want to configure.
[Optional] Select the Enable data collector name check box to enable the data collector to begin processing data.
Under LOGIN, select the login you want to configure from the list.
Under SAMPLE RATE, select one of the following options:
•
Recommended sample rate. The recommended sample rate is different for each data collector.
• Custom sample rate. The recommended sample rate appears, and you can adjust it to your system requirements.
[Sessions Data Collector only] Under SQL and Explain Thresholds, set the Elapsed
seconds and Total CPU seconds to determine whether the SQL and Explain text is collected for the session.
[Sessions Data Collector only] [Optional] Under New priority for lower priority
action, type one of the following:
•
$L
for low (default)
•
$M
for medium
•
$H
for high
•
$R
for rush
When the Lower Priority action executes for an alert, the session that met the alert rules has its priority changed to the above value.
Under DELETE DATA, do any of the following:
• Select the After check box to set the retention period using a number and calendar value.
• Select the Over check box, and then set the file size restriction using the Over box and list.
Click Apply to apply the settings.
Data Collector Rates
Teradata Viewpoint uses a sample rate to collect data from the Teradata Database system and uses a default retention rate to keep the collected data for a time period or up to a certain size. The sample rate and retention rate vary for each data collector and can be adjusted as needed.
Alert Request has no retention rate because this data collector does not store data.
Data Collector Name Recommended Sample Rate
Account Info
Alert Request
12 hours
5 minutes
Initial Sample
Rate
12 hours
5 minutes
Initial
Retention Rate
1 week
Not applicable
Teradata Viewpoint User Guide, Release 13.11
61
Chapter 7 System Administration
Data Collector Name Recommended Sample Rate
AWT Info
Database Space
Disk Space
Lock Info
Query Count
Resource Usage
Sessions
System Config
System Stats
Table Space
TASM Config
60 seconds
1 day
1 hour
1 hour
DBQLFlushRate value from the
DBS Control utility display command
Teradata Database logging sample rate from Teradata Performance
Monitor (PMON)
Teradata Database session sample rate from PMON
12 hours
Teradata Database resource sample rate from PMON
1 day
10 minutes
Initial Sample
Rate
60 seconds
1 day
1 hour
1 hour
1 Hour
1 hour
30 seconds
12 hours
30 seconds
1 day
10 minutes
Initial
Retention Rate
1 year
10 GB
1 year
10 GB
10 GB
10 GB
10 GB
1 week
1 year
TASM Distribution
TASM Exception
TASM State
TASM Summary
Virtual Storage
60 seconds
Sample rate returned by PM/API
TDWM Summary, otherwise 60 seconds
60 seconds
Sample rate returned by PM/API
TDWM Summary, otherwise 60 seconds
1 hour
60 seconds
60 seconds
60 seconds
60 seconds
1 hour
10 GB
None (both disabled)
1 year
1 year
1 year
1 year
1 year
Data Collector Catalog
Data collectors gather information from different sources and make the data available to
Teradata Viewpoint portlets.
Data Collector
Name
Description
Account Info Collects and retains available account strings from the
DBC.AccountInfo table. The Queries portlets use this data for the
Change Priority function.
Alert Request Monitors the dbcmngr.AlertRequest and dbcmngr.MonitorRequest tables on the Teradata Database system and collects data for further action by Teradata alerts.
Portlets Displaying Data
My Queries, Query Monitor
Alert Viewer
62 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Data Collector
Name
AWT Info
Description
Collects AMP worker task (AWT) related information through the AMP Load (ampload) Teradata Database utility.
Canary Queries Executes and retains user-defined canary queries and the System
Heartbeat canary query, which checks the responsiveness of a
Teradata Database system. This data is required for system health metrics.
Database Space Collects and retains database-space metrics from the
DBC.DataBaseSpace and DBC.DBase tables.
Disk Space Collects and retains disk-space and database-space metrics from the DiskSpace view of the DBC database.
Lock Info Collects and retains snapshots of lock information from Locking
Logger, a Teradata Database utility.
Query Count Collects and retains Teradata Database query counts and DBQL query log data recorded for each login from the tables
DBC.DBQLogTbl and DBC.DBQLSummaryTbl.
Resource Usage Collects and retains node and vproc resource usage data from the
DC.ResUsageSPMA table for node data and DC.ResUsageIPMA
table for vproc data.
Sessions Collects and retains Teradata Database session snapshots for each login from the Teradata Monitor partition by using the
Teradata Performance Monitor API. The Sessions collector retains utility data for the Query Monitor utility view.
Portlets Displaying Data
Capacity Heatmap, Metrics Analysis,
Metrics Graph, System Health
Canary Response Times, Productivity,
System Health
Capacity Heatmap, Metrics Graph,
Space Usage, System Health
Capacity Heatmap, Metrics Graph,
System Health
Lock Viewer
Capacity Heatmap, Productivity, Todays
Statistics, Metrics Graph
Metrics Analysis
System Config Collects and retains Teradata system configuration data used by
System Stats portlets and collectors. The collector retrieves the Teradata version from DBC.DBCInfo. To gather physical system configuration data, the collector also uses the MONITOR
PHYSICAL CONFIG request of the PM/API and executes
STATUS PDE of the Vproc Manager console partition utility.
This collector cannot be disabled.
Collects and retains system, node, and vproc statistics information from a Teradata Database. System statistics data is collected from the Teradata Monitor partition by using the PM/
API. Physical and virtual resource statistics are collected from
Teradata Database system memory. For Teradata Database systems earlier than Version 12.00, the System Statistics collector executes awtmon -s through the Teradata Console partition to collect AMP Worker Task (AWT) usage on MP-RAS systems.
AWT usage data is not available on Windows and Linux systems.
For Teradata Database Version 12.00 systems and later, the
System Statistics collector queries the Teradata Monitor partition to collect AWT usage information.
Capacity Heatmap, Metrics Graph, My
Queries, Query Monitor, Query
Spotlight, SQL Scratchpad, System
Health, Todays Statistics, Workload
Monitor
Alert Viewer, Canary Response Times,
Lock Viewer, Metrics Analysis, My
Queries, Node Resources, Productivity,
Query Monitor, Query Spotlight, Space
Usage, System Health, TVS Monitor,
Todays Statistics, Viewpoint
Monitoring; Workload Designer,
Workload Health, Workload Monitor;
Capacity Heatmap, Metrics Graph
Capacity Heatmap, Metrics Analysis,
Metrics Graph, Node Resources, System
Health, Todays Statistics, Workload
Health, Workload Monitor
Teradata Viewpoint User Guide, Release 13.11
63
Chapter 7 System Administration
Data Collector
Name
Table Space
Description
Collects and retains data about the space used by Teradata
Database system tables from the DBC.Dbase, DBC.TVM,
DBC.Indexes, and DBC.DataBaseSpace tables.
TASM Config Collects and retains all rule set information from the tdwm database.
TASM
Distribution
Collects information about the average percentage of CPU consumption used by each workload for each node in the
Teradata Database system. The TASM Distribution data collector executes schmon -M -p -P in the Teradata Console partition to collect information for the CPU Consumption pie chart graph in the Distribution view.
Portlets Displaying Data
Capacity Heatmap, Metrics Graph,
Space Usage, System Health
My Queries, Query Monitor, Workload
Health, Workload Monitor
Workload Monitor
TASM
Exception
TASM State
Collects and retains data from the DBC exception log.
Collects and retains system regulation data for Teradata
Database Version 12.00 and later. For Teradata Database systems earlier than Version 12.00, the Workload State collector gathers and retains period information.
Workload Monitor
Workload Monitor
TASM
Summary
Collects and retains active workload data from the tdwm database. Workload Health, Workload Monitor
Virtual Storage Collects and retains data on cylinder temperature and storage grade from space types managed by the Teradata Virtual
Allocation Manager (TVAM ) utility.
TVS Monitor
About Data Collector Status
The Teradata Systems portlet displays an icon when a data collector is disabled or functioning with errors. To see the most current status, it is recommended to refresh the system list and the data collector dialog box. You can view the icons in the system and data collector lists:
Error
Appears next to a Teradata Database system name or data collector name when one of these errors has been detected:
• Processing exception
• Incorrect Teradata System privileges
• Configuration errors, such as an incorrect password
• Teradata System is offline
Disabled
Appears next to a Teradata Database system name or data collector name if it is currently disabled.
64 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Viewing an Error Log
1
2
3
4
From the SYSTEMS list, click the Teradata Database system whose name is followed by
.
From the SETUP list, click Data Collectors.
The DATA COLLECTORS list refreshes the condition of each collector, identifying the collector that has the error.
Click next to the collector name of the log you want to view.
The Java stack trace log appears.
Click Close.
Configuring System Health Values
The System Health setup configures metrics and thresholds that are used by the System
Health and Productivity portlets. These thresholds are settings for the data collected by the
System Statistics, Sessions, Disk Space, and Canary Queries Collectors.
1
2
3
4
5
From the SYSTEMS list, click the Teradata Database system name.
From the SETUP list, click System Health.
Under THRESHOLDS, select one of the following options for each metric:
• Click Enabled to make the metric visible and use the threshold values to determine the system status.
• Click Disabled to ignore the metric and the threshold values.
• Click View Only to make the metric visible but to ignore the threshold values.
Adjust the degraded or critical thresholds for each metric.
Click Apply.
System Health Metrics
You can enable metrics and set thresholds that appear in the Productivity and System
Health portlets. Metrics available for selection and display are listed below.
Metric Description
CPU UTILIZATION Average node CPU use for User, System, and Wait IO
(CPU) USER
(CPU) SYSTEM
Average percentage of CPU time spent in user mode
Type
Percent
Percent
Average percentage of CPU time spent in system (kernel) mode Percent
(CPU) WAIT IO
NODE CPU SKEW
NODE IO SKEW
AMP CPU SKEW
AMP IO SKEW
Average percentage of CPU time spent waiting for I/O
Comparison of CPU use on the busiest node to the average node Percent
Comparison of disk use on the busiest node to the average node Percent
Comparison of CPU use on the busiest AMP to the average
AMP
Percent
Percent
Comparison of disk use on the busiest AMP to the average AMP Percent
Teradata Viewpoint User Guide, Release 13.11
65
Chapter 7 System Administration
Metric Description Type
AMP WORKER TASKS Average number of AMP worker tasks in use on each AMP Number
TOTAL DISK SPACE Percentage of total disk space currently in use Percent
Percent MAX DISK SPACE BY
AMP
COMPONENT
DOWN
Available disk space currently in use
Number of components, such as BYNETs or AMPs that are down
MEMORY USED Largest amount of memory used by a single node (KB)
MAX SPOOL BY AMP Available spool space in use
Number
Number
Percent
Number
Percent
MEMORY FAILURES Memory failures across all nodes
DBC DISK SPACE Available DBC disk space in use
ACTIVE SESSIONS Number of sessions with active queries
SYSTEM HEARTBEAT Canary query showing the response time of the most recent system heartbeat (ms)
Additional Canaries Additional canary queries that are defined for the system and used as part of the monitoring of system health
Number
Number
Number
Configuring Custom Status and Tooltips
You can change the default definitions of status and tooltips by enabling and defining custom status and tooltips for your system. These changes will appear in the System Health and Productivity portlets.
1
2
3
4
5
6
From the SYSTEMS list, click the Teradata Database system name.
From the SETUP list, click System Health.
Under Status and Tooltips, select Enable custom status.
Change the status and tooltip definitions for your system.
[Optional] Click at the end of the tooltip to restore the default definition.
Click Apply.
Copying System Health Settings
You can copy the custom status, tooltips, and threshold settings to other systems. You cannot copy canary queries to other systems. These changes will appear in the System
Health and Productivity portlets.
1
2
3
4
5
From the SYSTEMS list, click the Teradata Database system name.
From the SETUP list, click next to System Health.
Select Status and Tooltips to copy your custom status.
Select Thresholds to copy your threshold settings, except canary queries.
Select systems to copy your system health threshold settings.
66 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
6
Click OK.
Adding or Copying a Canary Query
After a Teradata System has been configured in Teradata Viewpoint, a canary query can be added to check the state of the system. If the system is active, the query returns a result.
1
2
3
4
5
6
7
8
9
10
11
12
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Canary Queries.
Do one of the following:
• To add a canary query, click .
• To copy a canary query, click in the row of the canary query you want to copy.
The canary query dialog box appears.
Enter a name for the canary query.
[Optional] Select the Enabled check box to enable the canary query.
Under SQL, enter or modify the query SQL code.
Under DEFAULT DATABASE, specify the default database.
Under LOGIN, select the login you want to configure from the list.
Under COLLECT DATA, specify the data collection frequency in seconds.
The recommended sample rate is 120 seconds. The retention rate is one year.
Under EXECUTE, select one of the following options:
• Click 24/7 to execute the query 24 hours a day, 7 days a week.
• Click Custom to execute the query during a time range by specifying the start time, end time, start day, and end day.
Under DELETE DATA, do any of the following:
• Select the After check box to set the retention period using a number and calendar value.
• Select the Over check box, and then set the file size restriction using the Over box and list.
Click Apply to apply the settings.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
Deleting a Canary Query
SYSTEM HEARTBEAT is a canary query that cannot be deleted. It is used to check whether the Teradata Database system is responsive.
1
2
3
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Canary Queries.
Click in the row of the canary query you want to delete.
Teradata Viewpoint User Guide, Release 13.11
67
Chapter 7 System Administration
68
4
A confirmation message appears.
Click OK.
The canary query row disappears.
About Alerts
Alerts monitor the performance on a Teradata Database system and automatically take action when events occur. You can add, copy, and configure alerts, as well as migrate
Teradata Manager alerts.
You can activate alert actions that send a notification, or take some other type of action, when a metric exceeds a threshold.
After you add alert action sets in the Alert Setup portlet, they appear in the Teradata
Systems portlet.
The types of alert actions you can choose are:
Send an Email
Choose a defined email address and text message. Event information, such as date, time, event name, threshold, and actual value, is automatically added to the body of the email message.
Run a Program
Choose a program to execute (.exe).
Run a BTEQ Script
Choose a BTEQ script.
Notify SNMP System
Choose an SNMP trap.
You can also choose alert actions for sessions including:
Abort Session
Abort the Teradata Database session for which an alert was detected. This action is only available for sessions.
Lower Priority
Set the priority of the Teradata Database session for which an alert was detected to the one specified in the New priority for lower priority action text box, located under Data Collectors setup for Sessions.
Alert Types
You can set alerts that take place when performance or database space events occur on one or more Teradata Database systems. Following is a list of alert types.
Alert Type
System
Description
Set an alert to occur when the performance on a Teradata Database reaches the specified threshold.
Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Alert Type
Node
Vproc
Database Space
Session
Canary Queries
System Health
Description
Set an alert to occur when the performance on a Teradata Database reaches the specified threshold at the node level.
Set an alert to occur when the performance on a Teradata Database reaches the specified threshold at the vproc level.
Set an alert to occur when the space used on any resource, such as an AMP, exceeds the specified percentage.
Set an alert to occur when the performance on a Teradata Database reaches the specified threshold at the session level.
Set an alert to occur when the response time for a canary query exceeds the threshold.
Set an alert to occur when the system health of a Teradata Database reaches the specified level.
Migrating Alerts
After a Teradata System has been configured in Teradata Viewpoint, all of the alerts except event alerts can be migrated from Teradata Manager to Teradata Viewpoint. Configured rates cannot be migrated, you must configure data collection rates in Teradata Viewpoint.
Alerts can only be migrated when there are no alerts in Teradata Viewpoint for a Teradata
Database system; therefore, migrate existing Teradata Manager alerts before adding and copying alerts.
1
2
3
4
5
6
7
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Alerts.
From the ALERT TYPES list, click an alert type.
Under ALERTS, click Migrate Alerts.
Enter a username and password that has permission to read data from the migrating database.
[Optional] Under SNMP Config, select the SNMP configuration assigned to any migrated action sets containing an SNMP action.
This is required when the alert actions being migrated contain any SNMP actions. The
SNMP configurations are defined in the Alert Setup portlet.
Click Apply.
A message appears in the dialog box with the number of alerts that were migrated.
Adding and Copying Alerts
After a Teradata System has been configured in Teradata Viewpoint, an alert can be added.
Alerts can only be migrated when there are no alerts in Teradata Viewpoint for a Teradata
Database system; therefore, migrate existing Teradata Manager alerts before adding and copying alerts.
1
From the SYSTEMS list, click the Teradata Database system name you want to update.
Teradata Viewpoint User Guide, Release 13.11
69
Chapter 7 System Administration
2
3
4
5
6
7
8
9
10
11
12
From the SETUP list, click Alerts.
From the ALERT TYPES list, click an alert type.
Do one of the following:
• To add an alert, click .
• To copy an alert, click in the row of the alert you want to copy.
The alert dialog box appears.
Enter a name for the alert.
[Optional] Select the Enabled check box to enable the alert.
Under Alert Rules, do the following:
a b
[Optional] Select databases to include or exclude for the Database Space alert type.
Specify All databases, All databases except, or Only include databases.
c a
Select a metric from the menus and enter a threshold value.
You also have the option to add or remove rules.
The alert triggers based on the value entered.
d
[Optional] Enter the number of minutes in the Only trigger if above is met for box.
The alert triggers if the rule is true for the amount of minutes you entered.
[Optional] Under Except these users, enter the name of the user for the Session alert type.
You also have the option to add or remove users.
Any session with the user is excluded from receiving the alert.
Under Alert Actions, do the following:
Select an alert action from the menu.
b
Enter the number of minutes in the Do not run twice in box.
The alert action does not run twice in the minutes you entered.
Under Severity, select a severity from the menu.
[Optional] Under Message, enter a message that appears when the alert action executes.
Click Apply to apply the settings.
The icon appears if the operation is successful. The icon appears if the operation fails. Verify that the settings are correct, and try again.
Deleting Alerts
An alert can be deleted any time after it has been added to a Teradata Database system.
1
2
3
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Alerts.
From the ALERT TYPES list, click an alert type.
70 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
4
5
6
Under ALERTS, select the alert name.
Click in the row of the alert you want to delete.
A confirmation message appears.
Click OK.
The alert row disappears.
Alert Metrics
The following table lists the metrics that you can specify in the system alert dialog box.
Metric Description
Node CPU Usage Average percent of CPU usage of all online nodes in the configuration
Type
Percent
Node Disk Usage Average percent of disk usage of all online nodes in the configuration Percent
Net A Usage Total BYNET utilization (average of the online BYNETs)
AMP CPU Usage Average percent of CPU usage of all online AMPs in the configuration
Percent
Percent
PE CPU Usage Average percent of CPU usage of all online PEs in the configuration Percent
Node CPU Skew Comparison of CPU use on the busiest node to the average node Percent
AMP CPU Skew Comparison of CPU use on the busiest AMP to the average AMP Percent
Total Sessions Total number of logged on sessions Number
Blocked Sessions Total number of blocked sessions Number
The following table lists the metrics that you can specify in the node and vproc alert dialog boxes.
Metric
Status
CPU Usage
Disk Usage
Net A Usage
CIC Usage
Description
Status of the node, where U means the resource is up and D means the resource is down
Percent of CPU usage that is not spent being active (not idle)
Percent of disk usage for this resource
Total BYNET utilization (average of the online BYNETs)
Percent of Channel Interface Controller usage for this resource
Disk Out Req Avg Average number of outstanding disk requests for this resource
Disk I/O Number of disk I/Os
Host I/O
Swaps
Mem Failures
Number of host I/Os
Total number of swap reads and swap writes
Segment allocation attempts that failed
Type
Character
Percent
Percent
Percent
Percent
Number
Number
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
71
Chapter 7 System Administration
Metric Description
Node CPU Usage Average percent of CPU usage of all online nodes in the configuration
Type
Percent
Node Disk Usage Average percent of disk usage of all online nodes in the configuration Percent
Net A Use Total BYNET utilization (BYNET receiver usage) Percent
Node CPU Skew Comparison of CPU use on the busiest node to the average node Percent
Total Sessions Number of sessions logged on to the PE
Available AWT Total available AMP worker tasks
Number
Number
Message Count Messages waiting on the vproc
AMP CPU Usage Average percent of CPU usage of all online AMPs in the configuration
Number
Percent
PE CPU Usage Average percent of CPU usage of all online PEs in the configuration Percent
AMP CPU Skew Comparison of CPU use on the busiest AMP to the average AMP Percent
The following table lists the metrics that you can specify in the database space alert dialog box.
Metric
Current Perm
Max %
Description
Current Perm % Percentage of the total permanent disk space the database is currently using (Current Perm divided by Max Perm)
Current Perm Max * Number of Vprocs / Max Perm
Type
Percent
Percent
The following table lists the metrics that you can specify in the session alert dialog box.
Metric
Host ID
Account
CPU Skew
CPU delta
AMP CPU
AMP I/O
Description
Host ID or LAN ID associated with the PE that processed the logon request for the session
Account from which a query was submitted
Type
Number
Character
CPU skew during the last sample Percent
Total CPU usage time consumed, in seconds, since the last sample Number
Number Current elapsed CPU time, in seconds, used on all AMPs by the associated session for executing requests
Current number of logical Reads and Writes issued across all
AMPs by the associated session
Number
Hot AMP CPU CPU time of the highest CPU utilized AMP during the collection interval
Connection Time How long the session has been connected
Number
Blocked Time
Idle Time
How long the query has been blocked
How long the query has been idle
Sessions Per User Number of sessions logged on under this username
Number
Number
Number
Number
72 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Metric
Partition
Description
Partition to which the session is connected (SQL, CONSOLE,
MONITOR)
Session from one of the Teradata utilities
Type
Character
Utility Session True or
False
The following table lists the metrics that you can specify in the canary query alert dialog box.
Metric Description
Response Time Number of milliseconds it took for the canary query to return a response
Type
Number
Setting Monitor Rates
You can set Teradata Database internal sample rates for sessions and resources. The sample rate you set becomes the recommended sample rate for each collector. Enabling the monitoring and logging process allows each collector to begin collecting data.
1
2
3
4
5
6
7
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Monitor Rates.
[Optional] Do one of the following:
• Select the check box to enable monitoring and logging process.
• Clear the check box to disable monitoring and logging process.
[Optional] Enter the Session Sampling sample rate in the text box.
The number you enter becomes the recommended sample rate for the Sessions collector.
[Optional] Select the Node/Vproc Sampling sample rate.
The number you choose becomes the recommended sample rate for the System
Statistics collector.
[Optional] Select the logging rate.
The number you choose becomes the recommended sample rate for the Resource Usage collector.
Click Apply to apply the settings.
Monitor Rates
After a Teradata System has been configured in Viewpoint, you can set Teradata Database internal sample rates for sessions, node/vproc, node logging, and vproc logging. These rates determine how often PM/API data are refreshed. Note the following when setting rates:
• The rates are based on the numbers set in Teradata Database.
• The sample rates for node/vproc, node logging, and vproc logging must be an integer divisor of 3600.
• The sample rate is valid when the new rate is an integer divisor of both current logging rates.
• The logging rate is valid when the new rate is an integer multiple of the current sample rate.
Teradata Viewpoint User Guide, Release 13.11
73
Chapter 7 System Administration
• The sample rate is applied before the logging rates are applied. When setting all three values at the same time, the sample rate must be a divisor of the old and the new logging rates. The logging rates must be a multiple of the old and the new sample rate.
Sampling Types
Sessions
Node/Vproc
Node logging
Vproc logging
Recommended Sample Rate Maximum Sample Rate
30 seconds 3600 seconds
30 seconds 3600 seconds
600 seconds
600 seconds
3600 seconds
3600 seconds
Configuring Log Table Clean Up
You can configure log table clean up to remove unnecessary information and reallocate space in the Teradata Database. After you enable the clean up process for a table and configure the retention period, you must set the time of the clean up schedule. The clean up process permanently deletes data and does not archive it.
1
2
3
4
5
6
7
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Log Table Clean Up.
From the TABLES list, click a table.
[Optional] Under Settings for table name , do one of the following:
• Select the Enabled check box to activate clean up and include the table in the clean up schedule.
• Clear the Enabled check box to disable clean up and exclude the table from the clean up schedule.
Under LOGIN, select the user you want to run the clean up task.
Under CLEAN UP, set the retention period using a number and calendar value.
Click Apply.
Clean Up Tables
The following clean up processes delete the associated system log tables:
Clean Up Process Name System Log Table Name
DBQ Log
• DBC.DBQLSummaryTbl
• DBC.DBQLStepTbl
• DBC.DBQLogTbl
• DBC.DBQLObjTbl
• DBC.DBQLExplainTbl
• DBC.DBQLSqlTbl
Resource Usage
• DBC.ResUsageSpma
• DBC.ResUsageSvpr
• DBC.ResUsageShst
74 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
Clean Up Process Name System Log Table Name
• DBC.ResUsageIpma
• DBC.ResUsageIvpr
• DBC.ResUsageScpu
• DBC.ResUsageSldv
• DBC.ResUsageSawt (V12 and later)
• DBC.ResUsageSps (V12 and later)
• DBC.ResUsageSpdsk (V12 and later)
• DBC.ResUsageSvdsk (V12 and later)
• DBC.ResUsageSobj (V2R6 only)
Event Log
• DBC.EventLog
TDWM
• DBC.TDWMSummaryLog
• DBC.TDWMEventLog
• DBC.TDWMExceptionLog
SW Event Log
• DBC.SW_Event_Log
Access Log
• DBC.AccLogTbl
Scheduling Clean Up
After the log table clean up has been configured, you can set a schedule. The schedule sets the time to delete all log tables that have been enabled for the clean up process. The clean up process permanently deletes data and does not archive it.
1
2
3
4
5
6
From the SYSTEMS list, click the Teradata Database system name you want to update.
From the SETUP list, click Clean Up Schedule.
[Optional] Under Clean Up Scheduler, do one of the following:
• Select the Enabled check box to activate the clean up schedule.
• Clear the Enabled check box to disable the clean up schedule.
Under LOGIN, select the user you want to collect the current time and Teradata
Database version.
Under DELETE TIME, enter a time at which the clean up task is run on the Viewpoint server by the DCS.
Current Teradata Time shows your information, or Unknown shows that a connection to Teradata Database cannot be made.
Click Apply.
Viewing Disk Usage
1
2
Click Disk usage.
[Optional] Mouse over the sparkline and graph to activate the information balloons and see detailed information.
Teradata Viewpoint User Guide, Release 13.11
75
Chapter 7 System Administration
3
4
[Optional] Click in a Teradata Database system box in the graph to view the disk usage for each data collector on that Teradata Database system.
[Optional] Click the collector name in the graph to see the corresponding configuration dialog box.
Disk Usage Graphics
The graphics on the Disk usage view show disk usage over time and the distribution of data on the Teradata Viewpoint server.
Graphics
Sparkline
Graph
Description
Shows 3 months of data. Information balloons show detailed data about a point in time.
Shows how the space used by the Teradata Viewpoint server is divided among Teradata Database systems and their data collectors. Mouse over or click Teradata Database system boxes in the graph to see detailed disk space and data collector information. If a system contains a region that is not associated with a collector, the information balloon displays
Unknown and the amount of data space.
User Manager
User Manager
The User Manager portlet allows the Teradata Viewpoint Administrator to view and manage
Teradata Viewpoint user accounts so that you can add or modify user accounts, assign roles, and reset passwords. You must have Teradata Viewpoint Administrator permission to access the User Manager portlet.
About the User List View
The User List view displays a list of all Teradata Viewpoint system users by FIRST NAME,
LAST NAME, PORTAL USER NAME, and EMAIL for each user. You can view, select, and search for individual user details or add Teradata Viewpoint user accounts. You can resize the column width or sort the contents of a column alphabetically.
Adding a User
If your Teradata Viewpoint portal is configured to use auto-provisioning, a user account is created automatically the first time a user logs on to Teradata Viewpoint. By default, autoprovisioned accounts are authenticated externally and assigned a default role. The authentication program (for example, LDAP) and default role are set during configuration.
1
2
3
4
From the User List view, click Add User.
Enter the portal username for the user account.
Enter the first name, last name, and email address for the user account.
Choose an authentication method:
76 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
5
6
7
• Select the Externally Authenticated check box to authenticate the selected user from an external source, such as LDAP.
• Clear the Externally Authenticated check box to authenticate the selected user from the Teradata Viewpoint internal database.
[Optional] In the Password field, enter a preliminary password for the user account.
Externally authenticated passwords are stored in the external source, such as LDAP, and validated from that source.
In the Re-enter Password field, enter the password again, exactly as it was entered previously.
Click Apply.
Searching for a User
You can search for an existing Teradata Viewpoint user, resize columns, and sort the information in the columns.
1
2
3
4
5
6
From the User List view, click the Find Users list.
Select a search method from the list.
In the contains field, enter the search criterion.
Click Find.
Users with information matching the search criterion are listed in the search results.
[Optional] Drag the column border in either direction to resize the column.
[Optional] Sort on a column by clicking the column header.
About the User Details View
The user details view allows you to delete a user, edit Teradata Viewpoint user accounts, and manage roles assigned to users.
The following tabs are available in this view:
General
Displays the selected user account, including name and email address.
Roles
Displays the available and assigned roles with role precedence for the selected user.
Assigning User Roles
A role must be defined in the Roles Manager portlet before it can be assigned to a user. The
Roles tab provides a list of available roles. Roles can be selected and assigned to the current user account.
You can assign roles to an existing Teradata Viewpoint user account. If your Teradata
Viewpoint portal is configured to use auto-provisioning, a user account is created automatically the first time a user logs on to Teradata Viewpoint. By default, autoprovisioned accounts are authenticated externally and assigned a default role. The authentication source (for example, LDAP) and default role are set during configuration.
Teradata Viewpoint User Guide, Release 13.11
77
Chapter 7 System Administration
1
2
3
4
5
6
From the User List view, browse the list of users or click Find Users to perform a search.
Click a user's name.
The General tab appears and displays information about the selected user.
Click the Roles tab.
Available roles are listed in the AVAILABLE PORTAL ROLES pane. Roles assigned to the user are shown in the ROLES FOR <User> pane.
Select a role from the AVAILABLE PORTAL ROLES list or select multiple roles by pressing Shift or Ctrl.
Click .
The selected roles appear in the ROLES FOR <User> pane. In the AVAILABLE
PORTAL ROLES pane, the assigned roles are dimmed.
Click Apply.
Setting Role Priority
You can set role priority for existing Teradata Viewpoint roles. Roles assigned higher priority take precedence over lower-priority roles.
1
2
3
4
5
6
From the User List view, browse the list of users or click Find Users to perform a search.
Click a user's name.
The General tab appears and displays information about the selected user.
Click the Roles tab.
Select a role from the ROLES FOR <User> pane to change the role's priority.
Change the role priority by doing one of the following:
• Click Up to increase the selected role priority. Roles at the top of the list have higher priority.
• Click Down to decrease the selected role priority. Roles at the bottom of the list are lower in priority.
Click Apply.
Editing a User
You can edit a Teradata Viewpoint user account, except the portal username. The username is defined when the user account is created using the User Manager portlet or it is stored in
LDAP when the user is externally authenticated.
1
2
3
4
From the User List view, browse the list of users or click Find Users to perform a search.
Click a user's name.
The General tab appears and displays information about the selected user.
[Optional] Change the first name, last name, or email address for the user account.
Choose an authentication method:
78 Teradata Viewpoint User Guide, Release 13.11
Chapter 7 System Administration
5
6
• Select the Externally Authenticated check box to authenticate the selected user from an external source, such as LDAP.
• Clear the Externally Authenticated check box to authenticate the selected user from the Teradata Viewpoint internal database.
[Optional] Change the password:
a b
Select the Change Password check box.
In the Password field, enter the password.
c
In the Re-enter Password field, enter the password exactly as it was entered previously.
If Externally Authenticated is selected, the Change Password check box and related password fields are unavailable, and the password must be changed in the external source.
Click Apply.
Changing a User Password
You can change a Teradata Viewpoint user password except if Externally Authenticated is selected. When the Change Password check box and related password fields are unavailable, and the password must be changed in the external source.
1
2
3
4
5
6
From the User List view, browse the list of users or click Find Users to perform a search.
Click a user's name.
The General tab appears and displays information about the selected user.
Select the Change Password check box.
The Password and Re-enter Password fields change color.
In the Password field, enter a password.
In the Re-enter Password field, enter the password again, exactly as it was entered previously.
Click Apply.
Deleting a User
You can delete a Teradata Viewpoint user account, except if you are logged on under the user account you want to delete.
1
2
3
4
From the User List view, browse the list of users or click Find Users to perform a search.
Click a user's name.
The General tab appears and displays information about the selected user.
Click Delete User.
A confirmation message appears.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
79
Chapter 7 System Administration
80 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 8
Alert Viewer
Alert Viewer
The Alert Viewer portlet allows users to view alerts defined for the system. The alert information in the summary view is updated every 30 seconds. Every alert has a time stamp, displaying the date and time at which the alert was issued.
You can filter the alerts by for example severity, time period, type, or name. You can also combine the filters to narrow the results further.
The ALERT DETAILS view displays detailed information about what triggered the alert, the source of the alert, and any relevant messages.
About Alerts
An alert is an event that the Teradata System Administrator defines as being significant. The
Teradata System Administrator assigns alert severity levels to rank alerts, and can also include an explanatory message. The severity levels are: critical, high, medium, or low. The alerts displayed in the Alert Viewer portlet are specific to your system.
Teradata Viewpoint User Guide, Release 13.11
81
Chapter 8 Alert Viewer
Alert Example
The Teradata System Administrator can define that a database exceeding a certain amount of storage usage triggers an alert. After it is exceeded, an alert appears in the Alert Viewer portlet with details to assist you in performing cleanup before the database becomes full.
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Filtering by Severity
You can filter the results to show only alerts with a specific severity level.
1
In the toolbar, click the button for the severity level you want to display.
Selecting a Time Period
You can filter the results to show alerts issued within the last 5 minutes, 1 hour, 1 day, or 1 week.
1
Select a time period from the list.
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
?
Wildcard or
Symbol Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Example Results
cat?
?cat
cats, catalog
scat, Scatter cat?l
cat???
catalog
catalog p???er
packer, parser, proper
82 Teradata Viewpoint User Guide, Release 13.11
Chapter 8 Alert Viewer
!
\
*
Wildcard or
Symbol
=
>
>=
<
<=
Description
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
Example Results
*cat cat*l cat*
*ews*er
*%
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
>60
>=60
<60
<=60
!cat33
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Teradata Viewpoint User Guide, Release 13.11
83
Chapter 8 Alert Viewer
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Viewing Alert Details
The ALERT DETAILS view displays details about a specific alert. The details include information about what triggered the alert, the source of the alert, and any relevant messages.
84 Teradata Viewpoint User Guide, Release 13.11
Chapter 8 Alert Viewer
1
In the summary view, click anywhere in a row to display the ALERT DETAILS view for that alert.
The ALERT DETAILS view appears. If the Source is Viewpoint, the following fields appear:
• Alert Type. The system action defines the alert type.
• Resulting Action. The Teradata Viewpoint Administrator defines the resulting action in
Teradata Systems Alerts.
Teradata Viewpoint User Guide, Release 13.11
85
Chapter 8 Alert Viewer
86 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 9
Calendar
Calendar
The Calendar portlet allows you to schedule and communicate events that might impact one or more Teradata Database systems. You must have appropriate permissions to manage system events. Unlike other portlets, only one instance of the Calendar portlet can be open in the Teradata Viewpoint portal. After the Calendar portlet is added to a portal page, the
Calendar menu option is no longer available on the Tools menu.
The Calendar portlet provides the following views:
Week View
The Week view displays all events between Sunday and Saturday of the selected week. In this view, only one day of the week is expanded at a time and event indicators on other days of the week are displayed at the scheduled time in the appropriate day column.
Month View
The Month view displays all events with start times falling between Sunday prior to the start of the selected month and Saturday after the end of the month. In this view,
event indicators are color-coded squares.
An event cannot last longer than 24 hours. You can define a time and priority to an event and, when you save it, an event indicator (a colored box) appears in the Week view or
Month view. A blue indicator means normal and a red indicates high priority events. Mouse over an event indicator to display details about the event. All times shown are for the time zone defined in the user profile.
About the Calendar Week View
The Week view displays all events falling between Sunday and Saturday of the selected week.
In this view, only one day of the week can be expanded at a time. The selected day can include one or more event summaries. The summary shows the event start and end times, and the event title. The remaining days are collapsed into narrow columns. Click in any day column to expand and display a different day of the week. Today's date is shaded darker if it falls within the selected week.
Teradata Viewpoint User Guide, Release 13.11
87
Chapter 9 Calendar
From the Week view, you can see event details by clicking anywhere in the event summary or event indicator. Events scheduled on other days of the week are indicated by a vertical line located at the scheduled time in the appropriate day column.
Add Event
Today
Opens the Add Event dialog box to add an event to the calendar.
Opens the weekly calendar that includes today.
Displays the Week view. If you click the date
(number), the selected day is expanded in the
Week view.
Displays the Month view.
Displays the previous week.
Displays the next week.
About the Calendar Month View
The Month view displays all events with start times that occur during the selected month.
Events shown in the selected month that start either in the previous or following month are indicated by a lighter color. Today is shaded darker if it falls within the selected month.
88 Teradata Viewpoint User Guide, Release 13.11
Chapter 9 Calendar
From the Month view, you can see event details by clicking on a date (number) or an event indicator.
Add Event
Today
Opens the Add Event dialog box to add an event to the calendar.
Opens the monthly calendar that includes today.
Displays the Week view. If you click the date
(number), the selected day is expanded in the
Week view.
Displays the Month view.
Displays the previous month.
Displays the next month.
About the Calendar Event Details View
The Event Details view displays the details of the selected event, including event priority, who created it (the author), whether it was modified, and when it is scheduled to occur. The event title and description are also included in this view.
Teradata Viewpoint User Guide, Release 13.11
89
Chapter 9 Calendar
From this view, you can manage the calendar and scheduled events.
Back
Edit
Delete
Return to the last displayed calendar view.
Open the Edit Event dialog box to edit the event details.
Delete this event.
Display the Week view containing the event.
Display the Month view containing the event.
Display the previous event.
Display the next event.
Adding an Event
Adding an event changes the system event calendar and affects all users. By default, new events are added to the currently selected day.
1
From either the Calendar Week or Month view, click Add Event
The ADD EVENT dialog box appears with the current date in the START and END date fields.
90 Teradata Viewpoint User Guide, Release 13.11
Chapter 9 Calendar
2
Set the START date:
a
Click inside the START date field.
A pop-up calendar appears.
3
4
5
6 b
Click or to navigate to the appropriate month.
c
Select the date from the pop-up calendar.
Set the END date.
Select the event START and END times from the list.
Select AM (morning) or PM (afternoon) from the list.
Select a priority:
•
Normal (default)
Teradata Viewpoint User Guide, Release 13.11
91
Chapter 9 Calendar
7
8
9
•
High
Enter the EVENT TITLE.
[Optional] Enter the event DESCRIPTION.
Click OK.
Editing an Event
Editing an event changes the event instance affecting all users.
1
2
3
From either the Calendar Week or Month view, select the event you want to edit.
The event details view appears.
Click Edit
The EDIT EVENT dialog box appears, showing event details.
Set the START date:
a
Click inside the START date field.
A pop-up calendar appears.
92
4
5
6
7 b
Click or to navigate to the appropriate month.
c
Select the date from the pop-up calendar.
Set the END date.
Select the event START and END times from the list.
Select AM (morning) or PM (afternoon) from the list.
Select a priority:
• Normal (default)
Teradata Viewpoint User Guide, Release 13.11
Chapter 9 Calendar
8
9
10
•
High
Enter the EVENT TITLE.
[Optional] Enter the event DESCRIPTION.
Click OK.
Deleting an Event
Deleting an event removes the event instance from the database and affects all users.
1
2
3
From either the Calendar Week or Month view, select the event you want to delete.
In the Event Details view, click Delete
A confirmation message appears.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
93
Chapter 9 Calendar
94 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 10
Canary Response Times
Canary Response Times
The Canary Response Times portlet allows you to analyze trends in single-system performance over a 60-minute or a 120-minute period using canary queries. Canary queries measure the time in milliseconds it takes for a user-defined query to run to completion. The same query is run repeatedly at a preselected interval to compare system workload over time. The longer the query takes to complete, the heavier the system workload at that time.
Canary response times can help you plan maintenance activities or schedule your workloads appropriately. For example, you can use this portlet to identify heavy workload periods on a
Teradata Database system so that you can avoid assigning tasks when system performance is marginal. You can also use canary response times to compare current performance with average performance measured across 1 or more weeks of historical data. Additionally, you can monitor and compare the performance of multiple Teradata Database systems by adding a Canary Response Times portlet to the portal page for each monitored system.
The summary view displays a graphic overview of selected canary query metrics monitored over the previous 60-minute or 120-minute period. Samples are taken once every minute, and a data point is added to the graph for each selected canary query metric.
Use the PREFERENCES view to select and organize canary query metrics to display, change thresholds, and adjust the vertical axis range for each metric. The Teradata Viewpoint
Administrator defines default metrics for the Canary Response Times portlet in the
Teradata Systems portlet. If enabled for your profile, you can set metrics that override the defaults.
The Canary Response Times portlet also gives you a means for comparing current and historical data by displaying past performance as an average of metric data points collected over time. The duration over which the averages are calculated can be changed using the
Past Averages tab.
About the Canary Response Times View
The Canary Response Times portlet monitors the workload of a single Teradata Database system over time. The summary view shows current and historical performance data using sparklines. The sparkline is a horizontal graph, showing the time (in milliseconds) a query takes to complete.
You can monitor the performance of multiple systems by adding a Canary Response Times portlet instance for each Teradata Database system you want to monitor. Canary response times, such as SYSTEM HEARTBEAT, are represented by a data point that is added to the
Teradata Viewpoint User Guide, Release 13.11
95
Chapter 10 Canary Response Times sparkline once every 60 seconds. Mouse over a sparkline to see an information balloon containing detailed information about the data point.
The sparkline types are:
Current Data
Shows the time (in milliseconds) that a query takes to complete. Each data point shown on the sparkline is the average of all response times recorded during a 120second interval, by default. The result is a curve sparkline. A large dot at the NOW point on the sparkline indicates the value of the last data point captured. A number to the right of the sparklines also indicates the NOW data-point value.
The Teradata Viewpoint Administrator sets the default data collection rate.
Past Average
Presents the averages for a user-selected number of weeks. The sparkline is shown as a skyline chart with flat, solid data points. Use this sparkline to compare the current performance from the previous 60-minute or 120-minute period with the system performance during the same 60-minute or 120-minute period 1 or more weeks in the past. Use the past-average data shown from NOW to 30 minutes or 60 minutes in the future to estimate system workload and enhance decision-making regarding workload assignments and resource allocation.
96
The summary views are:
Normal
The skyline shows 90 minutes of the average-workload data values from a userselected number of weeks in the past. The curve sparkline shows the actual data for the past 60 minutes using 1-minute data points.
Maximized
The skyline shows 180 minutes of the average-workload data values from a userselected number of weeks in the past. The curve sparkline shows the actual data for the past 120 minutes using 1-minute data points.
Toggle between the normal and maximized summary views using the (maximize) and
(minimize) portlet buttons.
Canary Response Times Metrics
The metrics available for display are canary queries.
Teradata Viewpoint User Guide, Release 13.11
Chapter 10 Canary Response Times
Metric
System Heartbeat
canary query name CQ
Default
Sparkline
Curve
Curve
Past Average
Line
Yes
Yes
Description
The predefined metric that is available for all systems. This is the default canary query.
Canary queries defined by the Teradata Viewpoint
Administrator appear on the Metrics tab.
About the Preferences View
The PREFERENCES view allows you to customize the portlet to monitor key metric trends that are important to efficient operation of your Teradata Database system. You can select only one system to monitor for each portlet instance.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
System
Select a Teradata Database system to monitor.
Metrics
Select and organize metrics to monitor in the summary view. Drag metrics from a list of all available metrics to the Preview pane where you can preview the results as you go.
Settings
Define settings for the metrics selected for display. Set thresholds to highlight metric values that are outside normal operating ranges. Set maximum vertical-axis ranges to limit the range of values displayed for selected metrics.
Past Averages
Adjust the number of weeks of data points used to calculate the average value for each displayed metric.
Selecting a System to Monitor
Use the System tab in the PREFERENCES view to select a Teradata Database system to monitor. Only one system can be selected for each portlet instance.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the System tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a system from the list.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
97
Chapter 10 Canary Response Times
Selecting Metrics to Monitor
Use the Metrics tab in the PREFERENCES view to select and organize the metrics shown in the summary view.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
The Preview pane uses sample data to show how metric rows are displayed in the summary view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do any of the following to change the way metric rows are displayed:
• Add a metric row. Drag a metric from the Select metrics for display list to the
Preview pane.
• Remove a metric row. Drag a metric from the Preview pane to the Metric Removal pane.
• Change metric row order. Drag a metric already in the Preview pane to a new location in the pane (up or down).
Changes to the metrics affect only the system currently selected, provided that the metrics are available on that system.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Metric Thresholds
Use the Settings tab in the PREFERENCES view to set the THRESHOLD and VERTICAL
AXIS RANGE values for the metrics selected for display. Only metrics selected using the
Metrics tab appear in the list of available metrics. Changes to the metrics affect only the system currently selected, provided that the metrics are available on that system.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Settings tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
[Optional] Enter the THRESHOLD setting for each available metric.
The threshold line does not appear in the view when the value entered is zero or blank.
[Optional] Enter the VERTICAL AXIS RANGE for each available metric.
The sparkline scales automatically when the value entered is zero or blank.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
98 Teradata Viewpoint User Guide, Release 13.11
Chapter 10 Canary Response Times
7
Click OK.
Setting Past Averages
Use the Past Averages tab in the PREFERENCES view to specify the number of weeks of data points used to calculate the average values displayed in metric graphs.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Past Averages tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter the number of weeks (1 to 99 weeks).
The default is 2 weeks.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
99
Chapter 10 Canary Response Times
100 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 11
Capacity Heatmap
Capacity Heatmap
The Capacity Heatmap portlet allows you to analyze resource usage trends on a Teradata
Database system to identify periods when a system is either over-utilized or under-utilized.
You can use this information to schedule resource-intensive jobs with minimal impact to other users.
The Capacity Heatmap portlet displays a grid with days on the horizontal axis and hours on the vertical axis. The metric and system are selected using selection menus. Time-range buttons on the toolbar allow you to select 1-month or 3-month periods. Each color-coded square represents an hour and provides a visual indication of the metric value during a 1hour period.
The PREFERENCES view allows you to set the work week and shift specifications.
About the Capacity Heatmap View
The CAPACITY HEATMAP view displays a graph of variations and trends in system resource usage. The graph plots days on the horizontal axis and hours along the vertical axis.
Customize the view using the menus, toolbar, and date-range slider to select metrics and date ranges.
The following list describes the features in this view:
Selection Menus
Shows the selections that define the metric currently being displayed. You can change the metric by changing the selections.
Toolbar
Contains the time-range buttons. Select a time-range button to change the graph.
You can use the Display and threshold buttons to define how the cells are shaded.
The time-range start and end dates are displayed on the toolbar next to the timerange buttons.
Date-Range Slider
Controls the amount of historical data displayed in the view and adjusts to display from 1 week to approximately 4 months of historical data. Dates outside the daterange slider timeline are shaded lighter. The darker-shaded area between the slider handles represents the current date range being displayed.
Teradata Viewpoint User Guide, Release 13.11
101
Chapter 11 Capacity Heatmap
Cell
Contains data about an hour on a specific date in the graph. Mouse over any cell to display an information balloon containing the date, hour, and metric value for the 1hour block.
Work Week and Shift Indicator
Highlights times when most users are normally accessing system resources with a box around the work week and work shift hours. Specify work day and shift hours using the PREFERENCES view.
102
Trend-Reporting Metrics
Database Metrics
Metric
Current Perm
Current Perm Max
Current Perm Min
Current Spool
Current Spool Max
Description
Sum of the current amount of permanent space (disk space) used across all vprocs
Maximum permanent space used on a single vproc
Minimum permanent space used on a single vproc
Sum of the current amount of spool space (disk space) used across all vprocs
Maximum spool space used on a single vproc
Type
Number
Number
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
Chapter 11 Capacity Heatmap
Metric
Current Spool Min
Current Temp
Current Temp Max
Current Temp Min
Peak Perm
Peak Perm Max
Peak Perm Min
Peak Spool
Peak Spool Max
Peak Spool Min
Peak Temp
Peak Temp Max
Peak Temp Min
Perm Limit
Spool Limit
Temp Limit
Description
Minimum spool space used on a single vproc
Sum of the current amount of temporary space (disk space) used across all vprocs
Maximum temporary space used on a single vproc
Minimum temporary space used on a single vproc
Sum of the largest amount of permanent space used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of permanent space used on a single vproc since the last reset
Type
Number
Number
Number
Number
Number
Number
Minimum of the largest amount of permanent space used on a single vproc since the last reset
Sum of the largest amount of spool used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of spool used on a single vproc since the last reset
Minimum of the largest amount of spool used on a single vproc since the last reset
Number
Number
Number
Number
Sum of the largest amount of temporary space used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of temporary space used on a single vproc since the last reset
Minimum of the largest amount of temporary space used on a single vproc since the last reset
Maximum space available for permanent space storage of all index tables, data tables, subtables, stored procedures, triggers, and permanent journals
Number
Number
Number
Number
Maximum space available for storage of spool database objects Number
Maximum space available for storage of temporary database objects
Number
Performance Metrics
Metric
Blocking Duration
Concurrency
Response Time
Retrieve Time
Rows Returned
Description
Total time the heartbeat query is blocked
Average number of active tasks running on the Teradata
Database while the heartbeat query is running
Type
Number
Number
Average response time of the selected canary query
Average time to retrieve all results
Number
Number
Average number of rows returned by the selected canary query Number
Teradata Viewpoint User Guide, Release 13.11
103
Chapter 11 Capacity Heatmap
Metric
Total Time
Description
Combined response time and retrieve time
Type
Number
Query Metrics
Metric
Abort Total
Active AMPs Avg
Active AMPs Max
Description
Number of aborted queries
Average number of active AMPs
Maximum number of active AMPs
Active AMPs Min
Active AMPs Stddev
Minimum number of active AMPs
Standard deviation of active AMPs
AMP CPU Skew Avg Average percent of AMP CPU skew
AMP CPU Skew Max Maximum percent of AMP CPU skew
AMP CPU Skew Min Minimum percent of AMP CPU skew
AMP CPU Skew Stddev Standard deviation percent of AMP CPU skew
AMP I/O Skew Avg
AMP I/O Skew Max
Average percent AMP I/O skew
Maximum percent AMP I/O skew
AMP I/O Skew Min Minimum percent AMP I/O skew
AMP I/O Skew Stddev Standard deviation of AMP I/O skew
Cache Total
Concurrency
Number of queries found in step cache
Average number of queries executing simultaneously during the hour
Delay Time Avg
Delay Time Max
Delay Time Min
Delay Time Stddev
First Response Time
Avg
First Response Time
Max
First Response Time
Min
First Response Time
Stddev
Max Steps in
Parallelism Avg
Max Steps in
Parallelism Max
Average query delay time
Maximum query delay time
Minimum query delay time
Standard deviation of query delay time
Average time it takes to receive the first response
Maximum time it takes to receive the first response
Minimum time it takes to receive the first response
Number
Number
Number
Number
Number
Number
Number
Standard deviation of time it takes to receive the first response Number
Average number of level-2 query steps executed in parallel Number
Maximum number of level-2 query steps executed in parallel Number
Type
Number
Number
Number
Number
Number
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Number
Number
104 Teradata Viewpoint User Guide, Release 13.11
Chapter 11 Capacity Heatmap
Metric
Max Steps in
Parallelism Min
Max Steps in
Parallelism Stddev
Description
Minimum number of level-2 query steps executed in parallel Number
Standard deviation of number of level-2 query steps executed in parallel
Number of Steps Avg Average number of query steps
Number of Steps Max Maximum number of query steps
Type
Number
Number
Number
Number of Steps Min Minimum number of query steps
Number of Steps Stddev Standard deviation of number of query steps
Query Count
Query Seconds Avg
Queries logged in dbc.QryLog, or the sum of Querycount from dbc.QryLogsummary during the sample period
Average query run time
Number
Number
Number
Query Seconds Max
Query Seconds Min
Query Seconds Stddev Standard deviation of query run time
Result Rows Avg Average number of result rows for the query
Result Rows Max
Result Rows Min
Result Rows Stddev
Spool Usage Avg
Spool Usage Max
Spool Usage Min
Spool Usage Stddev
Steps with Parallelism
Avg
Maximum query run time
Minimum query run time
Maximum number of result rows for the query
Minimum number of result rows for the query
Standard deviation of number of result rows for the query
Average amount of spool the query used
Maximum amount of spool the query used
Minimum amount of spool the query used
Standard deviation of amount of spool the query used
Average number of query steps with parallel steps
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Steps with Parallelism
Max
Steps with Parallelism
Min
Maximum number of query steps with parallel steps
Minimum number of query steps with parallel steps
Number
Number
Steps with Parallelism
Stddev
Standard deviation of number of query steps with parallel steps Number
Total CPU Time Avg Average CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Number
Total CPU Time Max Maximum CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Number
Teradata Viewpoint User Guide, Release 13.11
105
Chapter 11 Capacity Heatmap
Metric Description
Total CPU Time Min Minimum CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Total CPU Time Stddev Standard deviation of CPU time. CPU time is calculated as the total AMP CPU time plus the total parser and dispatcher CPU time for the query.
Total I/O Count Avg Average I/O count for the query
Total I/O Count Max Maximum I/O count for the query
Total I/O Count Min Minimum I/O count for the query
Total I/O Count Stddev Standard deviation of I/O count for the query
Warning Total Total number of queries with warnings
Type
Number
Number
Number
Number
Number
Number
Number
System Metrics
Metric
Active Sessions
AMP CPU Skew
AMP I/O Skew
Average Outstanding
Disk Reqs
AWT
Components Down
Description
Number of sessions with active queries
Outstanding disk requests (disk queue size)
Type
Number
Comparison of CPU use on the busiest AMP to the average AMP Percent
Comparison of disk use on the busiest AMP to the average AMP Percent
Number
CPU
Average number of AMP worker tasks in use on each AMP Number
Number of components, such as BYNETs or AMPs that are down
Average node CPU use. CPU is calculated as the sum of the user CPU, system CPU and wait I/O CPU usage percentages.
Number
Percent
CPU AWT Processing CPU resources spent processing an AMP worker task.
Normalized by multiplying the value returned from the
Teradata database by the number of physical CPUs on the associated node of the AWT.
CPU/Disk Ratio Ratio of CPU use to disk use. Calculated as the node CPU time divided by the physical disk usage.
CPU Dispatcher
Processing
DBC Disk Space
CPU resources spent in PE Dispatcher processing
Disk I/O
Disk Reads
Disk Usage
Disk Writes
Available DBC disk space in use Percent
Number of disk I/Os Number
Total physical disk reads per system during the sample period Number
Disk use on the system
Percent
Percent
Percent
Percent
Total physical disk writes per system during the sample period Number
106 Teradata Viewpoint User Guide, Release 13.11
Chapter 11 Capacity Heatmap
Metric
FSG Cache Miss
Host Block Reads
Host Block Writes
Index Ratio
Description
The percentage of FSG cache misses. Calculated as physical I/
Os divided by logical I/Os.
Total message blocks (one or more messages sent in one physical group) received from all clients
Total message blocks sent to all hosts
The percentage of index hits. Calculated as the index I/Os divided by the data block I/Os.
Logical MB/Sec Logical I/O of the system in megabytes per second
Max Disk Space By AMP Available disk space currently in use
Max Spool Space By
AMP
Memory Failures
Available spool space in use
Total memory failures across all nodes
Memory Resources
Net A Usage
Net Reads
Net Writes
Number
Percent
Percent
Total segments allocated to memory resources
Percent of BYNET A usage (BYNET receiver usage)
Number
Number
Percent
Total reads from the BYNET during the sample period Number
Total messages written to the BYNET during the sample period Number
Node CPU Skew
Node CPU Usage
Node I/O Skew
Nodes Down
Parallelism
Parser Usage
Comparison of CPU use on the busiest node to the average node Percent
Average percent of CPU usage of all online nodes in the configuration
Percent
Comparison of disk use on the busiest node to the average node Percent
Number of system nodes down Number
Percent Percentage of parallelism, calculated as the average CPU usage / maximum CPU usage x 100
Percent of CPU time the parser uses. It might exceed 100%.
Normalized by multiplying the value returned from Teradata
Database by the number of physical CPUs on the associated node of the parser.
Percent
PE CPU Usage Average CPU usage of the PEs. Calculated as the average usage of the parser and the dispatcher (both normalized values). Normalized by multiplying the value by the number of physical CPUs on the associated node of the PE.
Percent of I/O that are reads Read I/O
Session Login Count Sessions currently logged on to the system
Swap Drops Total pages or segments dropped from memory during the sample period due to swapping
Swap Reads Total pages or segments read into memory from the disk after a prior write or drop during the sample period
Type
Number
Number
Number
Number
Percent
Percent
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
107
Chapter 11 Capacity Heatmap
Metric
Swap Writes
System CPU
Total AMP CPU
Total Disk Space
Total Node CPU
User CPU
Vproc CPU Usage
Description
Total pages or segments written into swap area from memory during the sample period
Average CPU time spent in System mode
Total AMP CPU use
Total disk space currently in use
Total node CPU use
Average CPU time spent in User mode
Average percent of the CPU usage of all online vprocs in the configuration. Normalized by multiplying the value returned from Teradata Database by the number of physical CPUs in the associated vproc.
Average CPU time spent waiting for I/O
Type
Number
Percent
Percent
Percent
Percent
Percent
Percent
Percent Wait I/O CPU
Table Metrics
Metric
Index Count
Table Size
Description
Total number of indexes
Table size in bytes, including index, fallback, and journal subtables
Type
Number
Number
About the Capacity Heatmap Menus and Toolbar
The menus and toolbar allow you to specify the display attributes of the CAPACITY
HEATMAP view.
108
Use the selection menus to define a metric to monitor by selecting a:
•
System
•
Category
•
Metric
• Source or Variant
The first time you add the portlet to the page, the first available metric that you have permission to view is displayed. However, if you only have permission to view Table metrics, no data appears in the portlet until you select a metric using the selection menus.
The 1 mo and 3 mo buttons allow you to adjust the time range displayed. When clicked, 1 month or slightly more than 3 months of data up to the current end date appears in the view.
As you adjust the time range, the toolbar refreshes automatically to display the new time range in the toolbar.
Teradata Viewpoint User Guide, Release 13.11
Chapter 11 Capacity Heatmap
The Display button allows you to show or hide cell values and adjust high and low values that determine cell shading. The Display cell values check box is only available when the metric displayed is a percentage. If the selected time range results in narrow cells, cell values are not displayed even though the Display cell values check box is selected.
The threshold button displays the current threshold value and allows you to set the threshold slider and adjust the threshold. The threshold button and slider only appear when the metric displayed is a percentage.
If the threshold value is within the gray-scale range defined in the DISPLAY OPTIONS menu, the cells are displayed in a shade of the exception color equivalent to the corresponding gray-scale shade for that cell value.
Selecting Metrics and Systems to Monitor
You can use selection menus above the toolbar to choose a metric and system to monitor in the portlet, starting with the highest level menu on the far left. Different menus choices are available based on the previous menu, so not all menus apply to all metrics.
You are not required to make a selection from every menu each time you modify a metric selection.
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
6
7
8
9
Select a system from the menu.
Click Next.
Select a category from the menu.
Click Next.
Select a metric from the menu.
Click Next.
[Optional] Select a source or variant from the menu.
The menu for source or variant (sum, average, minimum, maximum, standard deviation of that metric) is only available for certain metric categories.
Click Next.
Adjusting the Date Range
Use the toolbar to set the range of dates to display for the selected metric and system.
1
Mouse over the date-range slider to activate the start-date and end-date handles.
Teradata Viewpoint User Guide, Release 13.11
109
Chapter 11 Capacity Heatmap
110
2
3
4
5
6
Drag the start-date handle until the intended start date is displayed in the toolbar.
Drag the end-date handle until the intended end date is displayed in the toolbar.
[Optional] Drag the date-range slider left or right to move the start-date and end-date handles in unison.
[Optional] Click 1 mo.
One month of historical data is displayed ending at the current end date.
[Optional] Click 3 mo.
Slightly more than three months of historical data is displayed ending at the current end date.
Setting Cell Values and Shading
The DISPLAY OPTIONS dialog box allows you to display cell values and to set metric thresholds for applying cell shading for each system monitored. The shading provides a visual overview by highlighting interesting values.
You must repeat this procedure for each available metric to ensure appropriate shading is applied when the metric is selected.
1
2
3
4
From the CAPACITY HEATMAP view, click Display.
The DISPLAY OPTIONS dialog box appears.
Do one of the following:
• Select the Display cell values check box to display cell values.
• Clear the Display cell values check box to display cells without values.
The Display cell values check box is only available when the metric displayed is a percentage. If the selected date range results in narrow cells, cell values are not displayed even though the Display cell values check box is selected.
Set the lower and upper threshold values in the Apply shading to values between boxes.
Click OK.
Values below the defined range appear in a very light shade of gray, while values above the range have a very dark shade. Values within the defined range appear in three shades of gray.
Setting the Threshold Value
The threshold button displays the current threshold value. Use the threshold button to access the threshold slider and adjust exception highlighting to show when the system is overutilized or under-utilized.
Teradata Viewpoint User Guide, Release 13.11
Chapter 11 Capacity Heatmap
The threshold button and slider only appear when the metric displayed is a percentage.
1
In the toolbar, click the threshold button.
The threshold slider appears.
2
3
4
Do one of the following:
• Click to set an upper threshold, highlighting all values greater than the selected value in shades of red.
• Click to set a lower threshold, highlighting all values less than the selected value in shades of gray.
Drag up or down to adjust the threshold value.
The CAPACITY HEATMAP view highlights change to reflect values above or below the threshold setting.
Click anywhere off the threshold slider to close the slider and return to the portlet.
About the Preferences View
The PREFERENCES view allows you to specify the work days and hours that are outlined with a border in the CAPACITY HEATMAP view. This setting helps to visually emphasize the days and hours during which most users are using system resources.
From the portlet frame, click to access the PREFERENCES view.
Configuring the Work Week and Shift
Use the Work week and shift tab in the PREFERENCES view to outline specific work days and hours with a dark gray border in the CAPACITY HEATMAP view. The work-week-andshift-border helps you easily identify the days and hours when system resources are used most.
1
2
3
From the portlet frame, click to access the PREFERENCES view.
Click the Work week and shift tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Teradata Viewpoint User Guide, Release 13.11
111
Chapter 11 Capacity Heatmap
4
5
6
7
Select the number of shifts (work days) per week from the Total number of shifts per
week list.
A single row displays for each shift.
For each shift, click the appropriate arrow, and then select the following:
•
Start day
•
Start time
•
Shift duration
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Example with Overlapping Shifts
When multiple shifts are defined on the same day and the hours you specify overlap, they are combined into one shift and are enclosed in one set of shift-indicator lines. The first shift starts at 6
AM and lasts for 8 hours, and the second shift starts at 1 PM and lasts for 8 hours. After you save these settings, the work-week lines and shift-indicator lines show a single shift from 6 AM to 9 PM.
The Work week and shift tab counts the combined Total number of shifts per week and the Shift
duration indicates a longer single shift.
112 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 12
External Content
External Content
The External Content portlet allows you to access your company browser-based tools from within the Teradata Viewpoint portal. Upon opening the External Content portlet, the web page appears, as set by the Teradata Viewpoint Administrator. If you have the appropriate permissions, you can use the PREFERENCES view to change the URL.
Caution:
Any valid web URL can be used. However, the portlet does not validate page content.
Accessing some URLs might display the page outside the portlet or redirect the portal page entirely.
About the Preferences View
The PREFERENCES view allows you to customize the URL that appears when the External
Content portlet is opened. You must have the appropriate permissions to change the URL.
Changing the URL
Use the PREFERENCES view to change the URL set by the Teradata Viewpoint Administrator.
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Enter the URL of your choice.
Press Enter.
A confirmation message appears.
Click Save.
The new URL is saved.
Setting the Default URL
Use the PREFERENCES view to change the default URL set by the Teradata Viewpoint
Administrator.
1
2
From the portlet frame, click to access the PREFERENCES view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
The URL reverts back to the default URL set by the Teradata Viewpoint Administrator.
Teradata Viewpoint User Guide, Release 13.11
113
Chapter 12 External Content
3
4
5
6
Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Enter the URL of your choice.
Click OK.
A confirmation message appears.
Click Save.
The new URL is saved as the default URL.
114 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 13
Lock Viewer
Lock Viewer
The Lock Viewer portlet allows you to view the lock data such as transaction identifiers, session identifiers, lock object identifiers, and global deadlocks. Viewing lock information helps you determine whether system performance has been degraded by a database lock.
Locate locks:
• In a specific database
• In a specific time frame
• Caused by a specific user
• That are blocking a specific user
About the Lock Viewer View
The LOCK VIEWER view displays a report containing information about the database locks that have occurred in the last 5 minutes, hour, day, or week. Lock data is displayed in the report only after the lock contention is resolved. Real-time lock contention data is not available. The Teradata Viewpoint Administrator must enable the appropriate collectors to display data in the LOCK VIEWER view.
The following list describes the features in this view:
Selection Menus
Shows the system and time frame for which the lock information is collected. You can choose a different system and time frame.
You can use the rewind feature with any time frame selection to retrieve lock data and compare it to data for a different date and time. You can only use the 5 minutes time frame with the rewind feature, and it is typically used to troubleshoot lock contentions.
Filters
Changes the displayed data to only show rows that match your filter criteria. You can use the filters to search for locks in a specific database, caused by a specific user, or that are blocking a specific user.
Lock Log Report
Shows lock data for transaction identifiers, session identifiers, lock object identifiers, and global deadlocks.
Teradata Viewpoint User Guide, Release 13.11
115
Chapter 13 Lock Viewer
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Export creates a .csv file containing the report information as shown.
116
Selecting a System and Time Frame
You can use selection menus above the filter to choose a system and time frame, starting with the highest level menu on the far left.
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
Select a system from the menu.
Click Next.
Select a time frame from the SELECT TIME FRAME dialog box.
Click Next.
The report appears in the summary view.
Viewing the Lock Log Report
Use the Lock Log Report to view the lock data such as transaction identifiers, session identifiers, lock object identifiers, and global deadlocks. Lock data is displayed in the report only after the lock contention is resolved. Real-time lock contention data is not available.
1
In the selection menu, choose a system and a time frame.
Teradata Viewpoint User Guide, Release 13.11
Chapter 13 Lock Viewer
2
The Lock Log report appears with the database locks that occurred within the period of time you specified.
[Optional] Drag the column border in either direction to resize the column.
Summary Locks
This topic describes the information that appears in the Lock Log Report.
Column
BLOCKING
HOST
BLOCKING
LEVEL
Description
Logical host ID of the transaction that imposed the lock
Lock level of the transaction that imposed the lock. The following values indicate the level:
• Database Locked at the database level
•
Table Locked at the table level
• RowHash Locked at the row level
•
RowRange Locked for a range of rows
Type
Number
Alpha
BLOCKING
MODE
BLOCKING
SESSION
Lock level mode of the transaction that imposed the lock.
The following values indicate the mode:
• Ac Access
•
Ex Exclusive
• Rd Read
•
Wr Write
Session number of the transaction that imposed the lock.
The following session numbers are used to indicate database internal sessions:
• 0 System User
• 1 System Accounting
• 2 System Recovery
• 3 Archive/Restore
Alpha
Number
BLOCKING USER Username who is blocking the query of another user
BLOCKTIME
DATABASE
Date and time at which the block occurred, displayed as MM/
DD/YY HH:MM:SS
Database name on which the lock was requested
DEADLOCK Indicates whether the lock contention results in a deadlock.
The following values indicate the type of deadlock:
• N No deadlock
•
Y Local deadlock
• G Global deadlock
Alpha
Number
Alpha or Number
Alpha
DELAY
HOST
Total time, in seconds, the transaction waited for the block Number
Logical host ID of the transaction that was waiting for the lock
Number
Teradata Viewpoint User Guide, Release 13.11
117
Chapter 13 Lock Viewer
Column
LEVEL
MODE
MULTIPLE
BLOCKER
Description
Lock level of the transaction that was waiting for the lock.
The following values indicate the level:
• Database Locked at the database level
• Table Locked at the table level
• RowHash Locked at the row level
• RowRange Locked for a range of rows
Lock level mode of the transaction that was waiting for the lock. The following values indicate the mode:
• Ac Access
•
Ex Exclusive
• Rd Read
• Wr Write
Type
Alpha
Alpha
Indicates whether more than one transaction encountered the same lock contention. The following values indicate the status:
• Y Two or more blocked transactions
•
N One blocked transaction
Alpha
PROCESSOR ID Unique identifier of the AMP where the lock was requested Number
SESSION Session number of the transaction that was locked. The following session numbers are used to indicate database internal sessions:
• 0 System User
• 1 System Accounting
• 2 System Recovery
Number
TABLE
USER
Table name on which the lock was requested
Username whose query is blocked by another user
Alpha or Number
Alpha or Number
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
118 Teradata Viewpoint User Guide, Release 13.11
Chapter 13 Lock Viewer
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
\
*
?
Wildcard or
Symbol
=
>
>=
<
Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
*cat cat*l cat*
*ews*er
*%
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
Example Results
cat?
?cat
cats, catalog cat?l
cat???
p???er
scat, Scatter
catalog
catalog
packer, parser, proper
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
>60
>=60
<60
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
Teradata Viewpoint User Guide, Release 13.11
119
Chapter 13 Lock Viewer
!
Wildcard or
Symbol
<=
Description
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
Example Results
<=60
!cat33
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Values exceeding the threshold are highlighted and displayed in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
120 Teradata Viewpoint User Guide, Release 13.11
Chapter 13 Lock Viewer
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Teradata Viewpoint User Guide, Release 13.11
121
Chapter 13 Lock Viewer
122 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 14
Metrics Analysis
Metrics Analysis
The Metrics Analysis portlet allows you to analyze resource usage trends on one or more
Teradata databases. Trends are graphed according to metrics such as CPU, memory, and throughput within a specified time range. The information is refreshed every 60 seconds.
The METRICS ANALYSIS view displays a graph with time on the horizontal axis, metric values on the vertical axis, and a different color for each metric sparkline. Sparklines show the average metric values as a line and have enhanced interactive features such as performance envelopes and information balloons that show the actual, minimum, and maximum metric values. Time-range buttons allow you to monitor current usage within the last hour or view usage trends over the last day, week, month, 3 months, or 6 months.
The PREFERENCES view allows you to select and organize metrics to display, set thresholds, and adjust the vertical axis range for each metric.
About the Metrics Analysis View
The METRICS ANALYSIS view displays a graph that represents Teradata Database resource usage. One or more Teradata databases can be monitored in the time range using the same or different metrics. Up to 10 metrics can be selected with different thresholds and vertical axis ranges. Each sparkline is assigned a color for every metric plotted in the graph.
The following list describes the features in this view:
Toolbar
Contains the time-range buttons. Select a time-range button to change the graph.
Graph
Plots the time range on the horizontal axis and the metric values on the vertical axis.
The sparkline is a colored line that represents the average metric values. Click a sparkline to see the highlighted sparkline and performance envelope. The performance envelope appears in a lighter shade of the sparkline color and represents the upper and lower metric values.
Time Range
Shows the range of time selected with the time-range button on the Toolbar. For example, if one hour is selected, times from the previous hour appear across the bottom of the graph.
Teradata Viewpoint User Guide, Release 13.11
123
Chapter 14 Metrics Analysis
Metrics
Lists the metrics that are defined in the PREFERENCES view. Each metric has a color and a checkbox. The metric name appears along with the name of the system being monitored if there is more than one system enabled. Select a metric name to see the sparkline highlighted and surrounded by the performance envelope in the graph.
124
About Metric Thresholds
Set thresholds to highlight metric values that are outside normal operating ranges. Set maximum vertical-axis ranges to limit the range of values displayed for selected metrics.
After a metric threshold is set in the PREFERENCES view, the threshold line is displayed across the graph with the threshold value to the right of the graph when you select a metric sparkline. As you mouse over the selected sparkline, information balloons display the data point value and the minimum and maximum performance envelope values. The sparkline and information balloons change to red when the threshold is exceeded during the time range. Threshold settings are optional and can be set any time after the metric has been configured.
Teradata Viewpoint User Guide, Release 13.11
Chapter 14 Metrics Analysis
Setting the Time Range
You can set the time range that is used to plot the graph.
1
In the toolbar, click a time-range button.
The portlet refreshes, and the graph is redrawn.
Removing Metrics from the Graph
You can disable metric sparklines.
1
Clear the color-coded checkbox on the metric line.
The metric sparkline disappears from the graph, but not from the list below the graph.
About the Toolbar
The toolbar allows you to choose the time range to display in the graph. The graph displays oldest data on the left and the most recent data on the right. For each metric, data is collected every 15 seconds and averaged according to the time range chosen.
Button
1 hour
1 day
1 week
1 mo
3 mo
6 mo
Description
Displays the metric values for the last hour, plotted by minute.
Displays the metric values for the last 24 hours, plotted by 15-minute periods.
Displays the metric values for the last week, plotted by hour.
Displays the metric values for the last month, plotted by 4-hour periods.
Displays the metric values for the last 3 months, plotted by 12-hour periods.
Displays the metric values for the last 6 months, plotted by day.
Clicking a different time-range button causes the portlet to immediately recalculate and update the graph based on the data points collected for the metrics.
Teradata Viewpoint User Guide, Release 13.11
125
Chapter 14 Metrics Analysis
Analysis Metrics
The following metrics are available to analyze resource usage.
System Metrics
Metric
Active Sessions
AMP CPU Skew
AMP I/O Skew
Average Outstanding
Disk Reqs
AWT
Components Down
Description
Number of sessions with active queries
Outstanding disk requests (disk queue size)
Type
Number
Comparison of CPU use on the busiest AMP to the average AMP Percent
Comparison of disk use on the busiest AMP to the average AMP Percent
Number
CPU
Average number of AMP worker tasks in use on each AMP Number
Number of components, such as BYNETs or AMPs that are down
Average node CPU use. Calculated as the sum of the user
CPU, system CPU, and wait I/O CPU usage percentages.
Number
Percent
CPU AWT Processing CPU resources spent processing an AMP worker task.
Normalized by multiplying the value returned from Teradata
Database by the number of physical CPUs on the associated node of the AWT.
CPU/Disk Ratio Ratio of CPU use to disk use. Calculated as the node CPU time divided by the physical disk usage.
CPU Dispatcher
Processing
DBC Disk Space
CPU resources spent in PE Dispatcher processing
Disk I/O
Disk Reads
Disk Usage
Disk Writes
FSG Cache Miss
Host Block Reads
Available DBC disk space in use Percent
Number of disk I/Os Number
Total physical disk reads per system during the sample period Number
Disk use on the system
Percent
Percent
Percent
Total physical disk writes per system during the sample period Number
Percentage of FSG cache misses. Calculated as the physical I/
Os divided by the logical I/Os.
Message blocks (one or more messages sent in one physical group) received from all clients
Percent
Number
Number
Host Block Writes
Index Ratio
Message blocks sent to all hosts
Index I/Os divided by the data block I/Os
Logical MB/Sec Logical I/O of the system in megabytes per second
Max Disk Space By AMP Available disk space currently in use
Max Spool Space By
AMP
Available spool space in use
Number
Number
Number
Percent
Percent
126 Teradata Viewpoint User Guide, Release 13.11
Chapter 14 Metrics Analysis
Metric
Memory Failures
Memory Resources
Net A Usage
Net Reads
Net Writes
Node CPU Skew
Node CPU Usage
Node I/O Skew
Nodes Down
Parallelism
Description
Memory failures across all nodes
Segments allocated to memory resources
Percent of BYNET A usage (BYNET receiver usage)
Reads from the BYNET during the sample period
Messages written to the BYNET during the sample period
Type
Number
Number
Percent
Number
Number
Comparison of CPU use on the busiest node to the average node Percent
Average percent of CPU usage of all online nodes in the configuration
Percent
Comparison of disk use on the busiest node to the average node Percent
Number of system nodes down
Percentage of parallelism, calculated as the average CPU usage / maximum CPU usage x 100
Number
Percent
Percent Parser Usage
PE CPU Usage
Percent of CPU time the parser uses. It might exceed 100%.
Normalized by multiplying the value returned from Teradata
Database by the number of physical CPUs on the associated node of the parser.
Average CPU usage of the PEs, calculated as the average usage of the parser and the dispatcher (both normalized values).
Normalized by multiplying the value by the number of physical CPUs on the associated node of the PE.
Read I/O Percent of I/O that are reads
Session Login Count Sessions currently logged on to the system
Swap Drops
Swap Reads
Pages or segments dropped from memory during the sample period due to swapping
Pages or segments read into memory from the disk after a prior write or drop during the sample period
Swap Writes Pages or segments written into swap area from memory during the sample period
Average CPU time spent in System mode System CPU
Total AMP CPU
Total Disk Space
Total Node CPU
User CPU
Vproc CPU Usage
Wait I/O CPU
Total AMP CPU use
Total disk space in use
Total node CPU use
Average CPU time spent in User mode
Average percent of the CPU usage of all online vprocs in the configuration
Average CPU time spent waiting for I/O
Percent
Percent
Number
Number
Number
Number
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Teradata Viewpoint User Guide, Release 13.11
127
Chapter 14 Metrics Analysis
Metric Colors
A fixed set of 10 colors is used for the sparkline, performance envelope, and information balloon. Each metric you select is listed below the graph, and its color is determined by the position selected in the PREFERENCES view.
5
6
3
4
7
8
9
10
Metric Position Color Example
1
2
Description
Green
Blue
Gold
Aqua
Pink
Gray
Lime
Brown
Teal
Purple
About the Preferences View
The PREFERENCES view allows you to customize the appearance of the graph.
In this view, you can:
• Add metrics to monitor and display.
• Delete metrics that you do not want to monitor or display.
• Select the ENABLE check box to plot the metric in the graph.
• Clear the ENABLE check box to remove the metric from the graph, but still list it below the graph.
• Reorder metrics to change their position and color.
• Select systems to monitor and display.
• Set a threshold that helps you track the performance of each metric.
• Set a vertical axis range to limit the maximum value to display.
From the portlet frame, click to access the PREFERENCES view.
Adding Metrics
You can add metrics that are plotted in the graph.
1
2
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
128 Teradata Viewpoint User Guide, Release 13.11
Chapter 14 Metrics Analysis
3
4
5
6
7
8
9
10
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Click to add a metric.
Up to 10 metrics can be listed.
Select a metric from the METRIC list.
[Optional] Select a system from the SYSTEM list.
The SYSTEM list is available when more than one Teradata Viewpoint server is enabled for the Data Collection Service.
[Optional] Enter a THRESHOLD value.
You can change the THRESHOLD settings for the metric at any time.
[Optional] Enter a VERTICAL AXIS RANGE value.
You can change the VERTICAL AXIS RANGE settings for the metric at any time.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Changing the Metrics Display Order
You can change the order of the metrics that are listed below the graph. Reordering the metric list affects which color is assigned to the metric.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
On the metric row, click and drag the row to its new location.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Enabling and Disabling Metrics
You can enable and disable metrics from the graph. When a metric is disabled, the metric sparkline is not plotted in the graph; however, data points are still being collected, so the metric is included in the list under the graph and can be reactivated at a later time.
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Select the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do one of the following:
Teradata Viewpoint User Guide, Release 13.11
129
Chapter 14 Metrics Analysis
5
6
• Clear ENABLE on the metric line to disable the metric.
• Select ENABLE on the metric line to enable the metric.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Deleting Metrics
You can delete metrics from the graph. When a metric is deleted in the PREFERENCES view, the metric sparkline is not plotted in the graph, and the metric is not included in the list under the graph.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Select the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Click on the metric row of the metric you want to delete.
The metric row disappears.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
130 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 15
Metrics Graph
Metrics Graph
The Metrics Graph portlet allows you to analyze resource usage trends on a Teradata
Database system within a specified time range to help identify daily or weekly periods when a system is either over-utilized or under-utilized. You can use this information to determine when to schedule resource-intensive jobs with minimal impact to other users.
The METRICS GRAPH view displays a graph with time on the horizontal axis and metric values on the vertical axis. The metric and system are selected using selection menus. Timerange buttons on the toolbar allow you to select 1-month or 3-month periods. The date range displayed can be adjusted dynamically using the date-range slider. The vertical scale for the metric values can be expanded or compressed by dragging the number up or down.
Each point on the graph represents the value collected for the selected metric during the preceding 1-hour period.
About the Metrics Graph View
The METRICS GRAPH view displays a graph of variations and trends in system resource usage. The graph plots the time range on the horizontal axis and the metric values on the vertical axis. Customize the view using the menus, toolbar, and date-range slider to select metrics and date ranges.
The following list describes the features in this view:
Selection Menus
Shows the selections that define the metric currently being displayed. You can change the metric by changing the selections.
Toolbar
Contains the time-range buttons. Select a time-range button to change the graph.
The time-range start and end dates are displayed on the toolbar next to the timerange buttons.
Date-Range Slider
Controls the amount of historical data displayed in the view and adjusts to display from 2 days to approximately 4 months of historical data. Dates outside the daterange slider timeline are shaded lighter. The darker-shaded area between the slider handles represents the current date range being displayed.
Teradata Viewpoint User Guide, Release 13.11
131
Chapter 15 Metrics Graph
Metric Value
Shows the numerical value of metric on the vertical axis. Mouse over a number and drag it up or down to expand or compress the vertical scale.
Sparkline
Displays metric data values in the graph. Mouse over a sparkline to see an information balloon containing the date and time the value was collected and the metric value.
132
Trend-Reporting Metrics
Database Metrics
Metric
Current Perm
Current Perm Max
Current Perm Min
Current Spool
Current Spool Max
Current Spool Min
Description
Sum of the current amount of permanent space (disk space) used across all vprocs
Maximum permanent space used on a single vproc
Minimum permanent space used on a single vproc
Sum of the current amount of spool space (disk space) used across all vprocs
Maximum spool space used on a single vproc
Minimum spool space used on a single vproc
Type
Number
Number
Number
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
Chapter 15 Metrics Graph
Metric
Current Temp
Current Temp Max
Current Temp Min
Peak Perm
Peak Perm Max
Peak Perm Min
Peak Spool
Peak Spool Max
Peak Spool Min
Peak Temp
Peak Temp Max
Peak Temp Min
Perm Limit
Spool Limit
Temp Limit
Description
Sum of the current amount of temporary space (disk space) used across all vprocs
Maximum temporary space used on a single vproc
Minimum temporary space used on a single vproc
Sum of the largest amount of permanent space used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of permanent space used on a single vproc since the last reset
Minimum of the largest amount of permanent space used on a single vproc since the last reset
Sum of the largest amount of spool used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of spool used on a single vproc since the last reset
Type
Number
Number
Number
Number
Number
Number
Number
Number
Minimum of the largest amount of spool used on a single vproc since the last reset
Sum of the largest amount of temporary space used across all vprocs since the last reset of the peak perm value
Maximum of the largest amount of temporary space used on a single vproc since the last reset
Minimum of the largest amount of temporary space used on a single vproc since the last reset
Number
Number
Number
Number
Maximum space available for permanent space storage of all index tables, data tables, subtables, stored procedures, triggers, and permanent journals
Number
Maximum space available for storage of spool database objects Number
Maximum space available for storage of temporary database objects
Number
Performance Metrics
Metric
Blocking Duration
Concurrency
Response Time
Retrieve Time
Rows Returned
Total Time
Description
Total time the heartbeat query is blocked
Type
Number
Number Average number of active tasks running on the Teradata
Database while the heartbeat query is running
Average response time of the selected canary query Number
Average time to retrieve all results Number
Average number of rows returned by the selected canary query Number
Combined response time and retrieve time Number
Teradata Viewpoint User Guide, Release 13.11
133
Chapter 15 Metrics Graph
Query Metrics
Metric
Abort Total
Active AMPs Avg
Active AMPs Max
Description
Number of aborted queries
Average number of active AMPs
Maximum number of active AMPs
Active AMPs Min
Active AMPs Stddev
Minimum number of active AMPs
Standard deviation of active AMPs
AMP CPU Skew Avg Average percent of AMP CPU skew
AMP CPU Skew Max Maximum percent of AMP CPU skew
AMP CPU Skew Min Minimum percent of AMP CPU skew
AMP CPU Skew Stddev Standard deviation percent of AMP CPU skew
AMP I/O Skew Avg
AMP I/O Skew Max
Average percent AMP I/O skew
Maximum percent AMP I/O skew
AMP I/O Skew Min Minimum percent AMP I/O skew
AMP I/O Skew Stddev Standard deviation of AMP I/O skew
Cache Total
Concurrency
Number of queries found in step cache
Average number of queries executing simultaneously during the hour
Delay Time Avg
Delay Time Max
Delay Time Min
Delay Time Stddev
First Response Time
Avg
First Response Time
Max
First Response Time
Min
First Response Time
Stddev
Average query delay time
Maximum query delay time
Minimum query delay time
Standard deviation of query delay time
Average time it takes to receive the first response
Maximum time it takes to receive the first response
Minimum time it takes to receive the first response
Number
Number
Number
Number
Number
Number
Number
Standard deviation of time it takes to receive the first response Number
Type
Number
Number
Number
Number
Number
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Percent
Number
Number
Max Steps in
Parallelism Avg
Max Steps in
Parallelism Max
Max Steps in
Parallelism Min
Average number of level-2 query steps executed in parallel
Maximum number of level-2 query steps executed in parallel Number
Minimum number of level-2 query steps executed in parallel
Number
Number
134 Teradata Viewpoint User Guide, Release 13.11
Chapter 15 Metrics Graph
Metric
Max Steps in
Parallelism Stddev
Description
Standard deviation of number of level-2 query steps executed in parallel
Number of Steps Avg Average number of query steps
Number of Steps Max Maximum number of query steps
Number of Steps Min Minimum number of query steps
Number of Steps Stddev Standard deviation of number of query steps
Query Count Queries logged in dbc.QryLog, or the sum of Querycount from dbc.QryLogsummary during the sample period
Query Seconds Avg
Query Seconds Max
Average query run time
Maximum query run time
Query Seconds Min Minimum query run time
Query Seconds Stddev Standard deviation of query run time
Result Rows Avg
Result Rows Max
Result Rows Min
Result Rows Stddev
Average number of result rows for the query
Maximum number of result rows for the query
Minimum number of result rows for the query
Standard deviation of number of result rows for the query
Spool Usage Avg
Spool Usage Max
Spool Usage Min
Spool Usage Stddev
Steps with Parallelism
Avg
Steps with Parallelism
Max
Steps with Parallelism
Min
Steps with Parallelism
Stddev
Average amount of spool the query used
Maximum amount of spool the query used
Minimum amount of spool the query used
Standard deviation of amount of spool the query used
Average number of query steps with parallel steps
Maximum number of query steps with parallel steps
Minimum number of query steps with parallel steps
Type
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Standard deviation of number of query steps with parallel steps Number
Number
Number
Number
Number
Number
Number
Number
Number
Total CPU Time Avg Average CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Total CPU Time Max Maximum CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Total CPU Time Min Minimum CPU time. CPU time is calculated as the total AMP
CPU time plus the total parser and dispatcher CPU time for the query.
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
135
Chapter 15 Metrics Graph
Metric Description
Total CPU Time Stddev Standard deviation of CPU time. CPU time is calculated as the total AMP CPU time plus the total parser and dispatcher CPU time for the query.
Total I/O Count Avg Average I/O count for the query
Total I/O Count Max Maximum I/O count for the query
Total I/O Count Min Minimum I/O count for the query
Total I/O Count Stddev Standard deviation of I/O count for the query
Warning Total Total number of queries with warnings
Type
Number
Number
Number
Number
Number
Number
System Metrics
Metric
Active Sessions
AMP CPU Skew
AMP I/O Skew
Average Outstanding
Disk Reqs
Description
Number of sessions with active queries
Type
Number
Comparison of CPU use on the busiest AMP to the average AMP Percent
Comparison of disk use on the busiest AMP to the average AMP Percent
Outstanding disk requests (disk queue size) Number
AWT
Components Down
Average number of AMP worker tasks in use on each AMP Number
Number of components, such as BYNETs or AMPs that are down
Number
Percent CPU Average node CPU use. CPU is calculated as the sum of the user CPU, system CPU and wait I/O CPU usage percentages.
CPU AWT Processing CPU resources spent processing an AMP worker task.
Normalized by multiplying the value returned from the
Teradata database by the number of physical CPUs on the associated node of the AWT.
Percent
CPU/Disk Ratio Ratio of CPU use to disk use. Calculated as the node CPU time divided by the physical disk usage.
CPU resources spent in PE Dispatcher processing
Percent
Percent CPU Dispatcher
Processing
DBC Disk Space
Disk I/O
Disk Reads
Disk Usage
Disk Writes
FSG Cache Miss
Host Block Reads
Available DBC disk space in use
Number of disk I/Os
Percent
Number
Total physical disk reads per system during the sample period Number
Disk use on the system Percent
Total physical disk writes per system during the sample period Number
The percentage of FSG cache misses. Calculated as physical I/
Os divided by logical I/Os.
Total message blocks (one or more messages sent in one physical group) received from all clients
Number
Number
136 Teradata Viewpoint User Guide, Release 13.11
Chapter 15 Metrics Graph
Metric
Host Block Writes
Index Ratio
Type
Number
Number
Logical MB/Sec Logical I/O of the system in megabytes per second
Max Disk Space By AMP Available disk space currently in use
Max Spool Space By
AMP
Memory Failures
Available spool space in use
Total memory failures across all nodes
Memory Resources
Net A Usage
Net Reads
Net Writes
Number
Percent
Percent
Total segments allocated to memory resources
Percent of BYNET A usage (BYNET receiver usage)
Number
Number
Percent
Total reads from the BYNET during the sample period Number
Total messages written to the BYNET during the sample period Number
Node CPU Skew
Node CPU Usage
Description
Total message blocks sent to all hosts
The percentage of index hits. Calculated as the index I/Os divided by the data block I/Os.
Node I/O Skew
Nodes Down
Parallelism
Parser Usage
Comparison of CPU use on the busiest node to the average node Percent
Average percent of CPU usage of all online nodes in the configuration
Percent
Comparison of disk use on the busiest node to the average node Percent
Number of system nodes down Number
Percent Percentage of parallelism, calculated as the average CPU usage / maximum CPU usage x 100
Percent of CPU time the parser uses. It might exceed 100%.
Normalized by multiplying the value returned from Teradata
Database by the number of physical CPUs on the associated node of the parser.
Percent
Percent PE CPU Usage
Swap Writes
Average CPU usage of the PEs. Calculated as the average usage of the parser and the dispatcher (both normalized values). Normalized by multiplying the value by the number of physical CPUs on the associated node of the PE.
Percent of I/O that are reads Read I/O
Session Login Count Sessions currently logged on to the system
Swap Drops Total pages or segments dropped from memory during the sample period due to swapping
Swap Reads Total pages or segments read into memory from the disk after a prior write or drop during the sample period
Total pages or segments written into swap area from memory during the sample period
System CPU
Total AMP CPU
Total Disk Space
Average CPU time spent in System mode
Total AMP CPU use
Total disk space currently in use
Percent
Number
Number
Number
Number
Percent
Percent
Percent
Teradata Viewpoint User Guide, Release 13.11
137
Chapter 15 Metrics Graph
Metric
Total Node CPU
User CPU
Vproc CPU Usage
Description
Total node CPU use
Average CPU time spent in User mode
Average percent of the CPU usage of all online vprocs in the configuration. Normalized by multiplying the value returned from Teradata Database by the number of physical CPUs in the associated vproc.
Average CPU time spent waiting for I/O
Type
Percent
Percent
Percent
Percent Wait I/O CPU
Table Metrics
Metric
Index Count
Table Size
Description
Total number of indexes
Table size in bytes, including index, fallback, and journal subtables
Type
Number
Number
About the Metrics Graph Menus and Toolbar
The menus and toolbar allow you to specify the display attributes of the METRICS GRAPH view.
138
Use the selection menus to define a metric to monitor by selecting a:
•
System
•
Category
•
Metric
• Source or Variant
The first time you add the portlet to the page, the first available metric that you have permission to view is displayed. However, if you only have permission to view Table metrics, no data appears in the portlet until you select a metric using the selection menus.
The 1 mo and 3 mo buttons allow you to adjust the time range displayed. When clicked, 1 month or 3 months of data up to the current end date appears in the view. As you adjust the time range, the toolbar refreshes automatically to display the new time range in the toolbar.
Selecting Metrics and Systems to Monitor
You can use selection menus above the toolbar to choose a metric and system to monitor in the portlet, starting with the highest level menu on the far left. Different menus choices are available based on the previous menu, so not all menus apply to all metrics.
You are not required to make a selection from every menu each time you modify a metric selection.
Teradata Viewpoint User Guide, Release 13.11
Chapter 15 Metrics Graph
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
6
7
8
9
Select a system from the menu.
Click Next.
Select a category from the menu.
Click Next.
Select a metric from the menu.
Click Next.
[Optional] Select a source or variant from the menu.
The menu for source or variant (sum, average, minimum, maximum, standard deviation of that metric) is only available for certain metric categories.
Click Next.
Adjusting the Date Range
Use the toolbar to set the range of dates to display for the selected metric and system.
1
Mouse over the date-range slider to activate the start-date and end-date handles.
2
3
4
5
6
Drag the start-date handle until the intended start date is displayed in the toolbar.
Drag the end-date handle until the intended end date is displayed in the toolbar.
[Optional] Drag the date-range slider left or right to move the start-date and end-date handles in unison.
[Optional] Click 1 mo.
One month of historical data is displayed ending at the current end date.
[Optional] Click 3 mo.
Three months of historical data is displayed ending at the current end date.
Teradata Viewpoint User Guide, Release 13.11
139
Chapter 15 Metrics Graph
140 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 16
My Queries
My Queries
The My Queries portlet allows you to view and manage your queries across multiple
Teradata Database systems. You can see if queries are queued or blocked, and you can see their impact on system resources.
Use the My Queries portlet to view information about queries in either the summary view or the details view. The summary view contains a table with one row allocated to each of the sessions logged on under one or more user names. Select a row in the summary view to see additional session and query information in the details view. Use the SQL, Explain, Blocked
By, or Query Band tab in the details view to display information for the selected session.
The PREFERENCES view allows you to select one or more Teradata Database systems, and then select one or more users per system to monitor. From this view, you can also select a format for the SQL that appears in the query details view.
About the My Queries View
After you add an account in the Profile portlet for each Teradata Database system you want to monitor, the MY QUERIES view displays queries running on the selected systems under your logon.
Filters
Filters allow you to change displayed data by showing only rows that match your filter criteria. You can also sort on the column headers to find the queries you want.
Sessions
The MY QUERIES view displays information about each session in a table, with columns configured specifically for the current view. These statistics are sampled and the table is refreshed every 30 seconds. Highlighted values in any row indicate that a session has exceeded threshold criteria for a specific metric. Click anywhere in a row to see session details.
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Export creates a .csv file containing the report information as shown.
Teradata Viewpoint User Guide, Release 13.11
141
Chapter 16 My Queries
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
*
?
Wildcard or
Symbol Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
cat?
?cat
cat?l
cat???
p???er
Example Results
cats, catalog
scat, Scatter
catalog
catalog
packer, parser, proper
*cat cat*l cat*
*ews*er
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper
142 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
!
\
Wildcard or
Symbol
=
>
>=
<
<=
Description
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
*%
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
Example Results
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
>60
>=60
<60
<=60
!cat33
what is 100%?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
Teradata Viewpoint User Guide, Release 13.11
143
Chapter 16 My Queries
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Values exceeding the threshold are highlighted and displayed in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Summary View Metrics
Metrics appearing in the summary view can be selected using the Columns dialog box.
Metrics available for monitoring and display are listed below.
Metric
ACCOUNT
BLOCKED TIME
Description
Account from which a query was submitted
How long the query has been blocked
Type
Character
Number
144 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
Metric
CPU SKEW
CPU USE
Delta CPU
Delta I/O
DURATION
HOST
Description
CPU skew during the last sample
Type
Percent
Percent of the total amount of available CPU used by the query Percent
Number Total CPU usage time consumed, in seconds, since the last sample
I/O count since the last sample
How long the query has been running
Host ID or LAN ID associated with the PE that processed the logon request for the session
Number
Number
Number
Number IMPACT CPU
IN STATE
I/O SKEW
PARTITION
PJI
CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
How long the query has been in the current state
I/O skew during the last sample
Partition in which the query is running
QUERY BAND
REQ CPU
REQ I/O
SESSION ID
SPOOL
START
STATE ICON
STATE
SYSTEM
Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
Entire query band string (query bands are a set of name-value pairs defined by the user to tag sessions or transactions with an ID through an SQL interface)
Number of CPU seconds needed to run the query
Number of disk I/Os performed to run the query
Unique session identifier
Amount of spool space the query requires
Time that the query started running on Teradata Database
Icon representing the current state of the query
Text describing the current state of the query
Full name of the system running the query
TEMP SPACE Amount of temp space used by the query
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
USERNAME
WORKLOAD
Name of the user who submitted the query
Workload in which the query is running
Number
Percent
Character
Number
Character
Character
Character
Number
Number
Number
Number
Number
Icon
Character
Character
Number
Number
Symbol Reference
The following symbols appear in the portlets:
Teradata Viewpoint User Guide, Release 13.11
145
Chapter 16 My Queries
Symbol Name Definition
ABORTING Query has been aborted, and changes are being rolled back.
ACTIVE
BLOCKED
DELAYED
IDLE
Query is running.
Query is waiting for a lock held by another query.
Query is in a delay queue waiting to run.
No query is running.
OTHER
PARSING
Query is in an unknown state.
Query is being parsed before running.
QTDELAYED Query is waiting for rows to be inserted into a queue table.
RESPONSE Query is returning results to the user.
SESDELAYED Query is delayed because a utility limit has been exceeded in FastLoad,
MultiLoad, FastExport, or ARC.
About the Details View
The details view displays statistics and information about the selected session. This view can be accessed by clicking on a session row in the summary view.
When viewing a request, you can see detailed read-only information from the following tabs:
Overview
Key statistics for a session. Any value exceeding preset thresholds is highlighted.
SQL
SQL for the selected query.
Explain
Explain steps for the query, including step statistics and Explain text.
Blocked By
Details about other queries that are blocking this query.
Query Band
Displays the query band name and value for the selected query.
Delay
Details about rules delaying this query.
Use the and buttons to move through sessions without returning to the summary view.
146 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
Details View Metrics
Query Information
Metric
STATE
CPU USE
CPU
CPU Δ
Description
Query state, such as ACTIVE, BLOCKED, TERMINATE
Type
Character
Number TIME IN STATE
TOTAL DURATION
I/O
How long the query has been in the current state, displayed as
HH:MM:SS
How long the query has run, displayed as HH:MM:SS
Percent of the total amount of available CPU used by the query Percent
Total CPU usage time, in seconds, consumed by the query Number
Total CPU usage time consumed, in seconds, since the last sample
Number of I/O operations the query performed
Number
Number
Number
I/O Δ
CLASSIFICATION
MODE
I/O count since the last sample
Indicator that a query or session matches a workload definition (WD) if Teradata DWM Category 3 is enabled.
Available values are:
Number
Character
CPU SKEW
AUTO = Workload was assigned automatically by TASM
REQUEST = Workload was assigned manually at the
Request Level
SESSION = Workload was assigned manually at the
Session Level
This field is empty if this is not a DBC/SQL session or if
Teradata DWM Category 3 is disabled.
CPU skew during the last sample Percent
Teradata Viewpoint User Guide, Release 13.11
147
Chapter 16 My Queries
Metric
I/O SKEW
SPOOL
Description
I/O skew during the last sample
Amount of spool space the query is using
TEMP SPACE
WORKLOAD
Amount of temp space the query is using
Workload where the query is actively running
PJI Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
IMPACT CPU CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
Type
Percent
Number
Number
Character
Number
Number
Number
Session Information
Metric
USER
ACCOUNT
SOURCE
PARTITION
REQUESTS
Description
Name of the user that submitted the query
Account of the user that submitted the query
Source details, such as application name, IP address, and host user name
Partition in which the query is running
Number of queries submitted by the session
Type
Character
Character
Character
Character
Number
About the Overview Tab
The Overview tab provides detailed information about key metrics for the selected session and its queries. The metric values displayed provide a snapshot of Teradata Database system status. Metrics that exceed defined thresholds are highlighted.
About the SQL Tab
The SQL tab displays the SQL for the selected query. This information is read-only. Use to export the SQL as a .txt file.
About the Explain Tab
The Explain tab displays an abbreviated version of the Step statistics and the Explain text that result from an Explain request in an SQL session. Each Explain step is uniquely identified with a number where the background of the number box indicates status. This information is read-only. This tab only appears when a query meets certain thresholds established by the Teradata Viewpoint Administrator. Use to export the Explain data as a .csv file.
148 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
Step Information
Step Number
Description
• Completed steps are at the top of the list and indicated by a black number box.
• Active steps are indicated by a pulsating number box (flashes gray and white).
• Steps to run are at the bottom of the list and indicated by a white number box.
Confidence Level Indicator Icon
EST. TIME
EST. ROWS
ACTUAL TIME
ACTUAL ROWS
- No confidence in the estimate
- Low confidence in the estimate
- High confidence in the estimate
- High confidence in the estimate due to a join index
Estimated execution time for the step.
Estimated number of rows for the step.
Actual CPU time consumed by the step, or blank if the step has not run.
Actual number of rows for the step, or blank if the step has not run.
About the Blocked By Tab
The Blocked By tab displays details about other queries that are blocking this query. This information is read-only. This tab is only available when the selected query is blocked.
Blocked By Information
USERNAME
HOST
SESSION ID
LOCK TYPE
STATUS
LOCKED
Description
Name of the user that is running the query that holds the lock.
Host ID or LAN ID associated with the PE that processed the logon request for the session.
Session ID of the blocking query.
Type of lock. Type can be Exclusive, Read, Write, or Access.
Lock status. Status can be Waiting or Granted.
Name of the locked object.
About the Query Band Tab
The Query Band tab displays the session and transaction query bands for the selected query.
This information is read-only.
Query Band Information
NAME
VALUE
Description
The name of the query band.
The value of the query band.
Teradata Viewpoint User Guide, Release 13.11
149
Chapter 16 My Queries
About the Delay Tab
The Delay tab displays details about rules that are delaying this query. The Delay tab is available if the system being monitored is a Teradata Database 13.10 system or newer. The information on the Delay tab is read-only, and no actions are available from this tab. This tab is available when the selected query is delayed. The Delay tab displays all rules that apply to the query and a scroll bar appears if there are more than two rules.
Delay Information
BLOCKING COUNT
Description
Number of consecutive times this session has blocked at least one other session.
Amount of time request has been delayed by a utility throttle rule.
UTILITY THROTTLE DELAY
TIME
SYSTEM THROTTLE DELAY
TIME
WORKLOAD THROTTLE
DELAY TIME
RULE NAME
RULE TYPE
OVERRIDABLE
Amount of time request has been delayed by a system throttle rule.
Amount of time request has been delayed by a workload throttle rule.
Name of rule causing request delay.
TASM type of rule causing request delay.
Indicates if the Teradata Database Administrator can abort or release the request.
About Managing Queries and Sessions
You can manage queries and sessions to improve workload performance.
Abort
Abort the selected query or session.
Change Priority
Change the priority of the selected query or session.
Change Workload
Change the workload of the selected query or session.
Release Query
Release the selected query from a delay queue.
You must log on with a user ID that has permissions to abort, change priorities or workloads, or release queries. If you log out, close a portlet, or open a new portlet instance, you must log on again.
Note: Change Workload is available only if Teradata Active System Management (TASM) is enabled. If TASM is disabled, Change Priority is available. If you do not see Change
Workload or Change Priority in the menu, the system you are monitoring does not support these features or you do not have permission to use them. If the query you are monitoring is delayed, only Release Query is available.
150 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
Aborting a Query or a Session
You can abort a query or a session that is blocking other queries or consuming too many resources.
1
2
3
4
5
6
7
8
From the summary view, click the row of the query you want to abort.
Click .
Select Abort.
[Optional] Log on.
Click Next.
Select one of the following:
• Abort Query to abort the selected query.
• Abort Session to abort the selected query and log off the session.
Click Next.
Click Next to confirm your selection.
Changing the Priority of a Query or Session
You can change the priority of a query or session to allow higher priority queries to run or to balance session resources.
This menu item is only available when workloads are not enabled and the system being monitored supports this feature.
1
2
3
From the summary view, click the row of the query with a priority that needs to change.
Click .
Select Change Priority.
4
5
6
7
8
9
10
11
[Optional] Log on.
Click Next.
From the list of accounts that have been assigned to the user, select an account to copy the account name to the Edit new account string box.
[Optional] Edit the account string or enter a new account string.
Click Next.
Select one of the following:
•
Move just this query to change the priority of the selected query.
• Move this query and all future queries associated with this session to change the priority of the selected query and all subsequent queries in the current session.
Click Next.
Click Next to confirm your selection.
Teradata Viewpoint User Guide, Release 13.11
151
Chapter 16 My Queries
Changing the Workload of a Query or Session
You can change the workload of a query or session to either allow higher priority workloads to run or to balance workload resources.
This menu item is only available when workloads are enabled and the system being monitored supports this feature.
1
2
3
4
5
6
7
8
9
10
From the summary view, click the row of the query with the workload that needs to change.
Click .
Select Change Workload.
[Optional] Log on.
Click Next.
[Optional] Select a different workload from the list.
Click Next to confirm your selection.
Select one of the following:
• Move just this query to change the workload of the selected query.
•
Move this query and all future queries associated with this session to change the workload of the selected query and all subsequent queries in the same session.
Click Next.
Click Next to confirm your selection.
Releasing a Query
You can release a query that is in the delay queue and waiting to run.
1
2
3
4
5
6
From the summary view, click the row of the query you want to release.
Click .
Select Release Query.
[Optional] Log on.
Click Next.
Click Next to confirm your selection.
Changing Multiple Sessions
You can change system resources for multiple sessions, users, or account strings in the following locations:
• All sessions table from the summary view
• User sessions table from the details view
• Account strings table from the details view
1
2
Click on a session from the summary view.
[Optional] Access user or account string tables from the details view.
152 Teradata Viewpoint User Guide, Release 13.11
Chapter 16 My Queries
3
4
5
6
7 a
Click on a session.
b
Click the USER or ACCOUNT link under SESSION INFO.
Select a session action from the Display Configuration menu.
Check boxes appear next to the sessions.
Do one of the following:
• Select the check box in the column heading to select all sessions.
• Select the check boxes for specific sessions.
Select Next.
[Optional] Log on.
Follow the instructions on subsequent screens.
About the Preferences View
The PREFERENCES view allows you to select systems and users to monitor and to select a format for the SQL that appears in the query details view.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
Systems
Select the Teradata Database systems and login user names to monitor.
Display
Choose formatted or unformatted SQL to display on the SQL tab of the query details view.
Selecting Systems and Users to Monitor
Use the Systems tab in the PREFERENCES view to select Teradata Database systems and login user names to monitor.
Define user names for each system using the Profile portlet. Only predefined user names appear on the Systems tab.
1
2
3
From the portlet frame, click to access the PREFERENCES view.
Click the Systems tab.
Click to expand system names and show available user names.
Teradata Viewpoint User Guide, Release 13.11
153
Chapter 16 My Queries
154
4
5
6
7
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
For each Teradata Database system and user name you want to monitor:
a
Select the systems on which you want to monitor queries.
b
Select the user names for each system that you want to monitor.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Selecting Display Options
The Display tab in the PREFERENCES view allows you to format the SQL that appears on the SQL tab of the query details view.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Display tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do one of the following:
• Select the Format SQL check box to use formatted SQL.
• Clear the Format SQL check box to use unformatted SQL.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
CHAPTER 17
Node Resources
Node Resources
The Node Resources portlet allows you to monitor physical and virtual resources to locate over or under-utilized resources and isolate performance or system issues. You choose which system and resources to monitor on your Teradata Database system.
The Node Resources portlet provides metrics about:
• Percentage of CPU used by nodes or vprocs
• How system resource usage is spread across the vprocs
• How much physical disk I/O, BYNET traffic, or host reads and writes are occurring
• Whether congestion or excessive swapping is an issue on a single or group of nodes or vprocs
These metrics allow you to analyze data and detect issues such as poor node or vproc parallel efficiency, or higher than expected disk read/write ratio or swap I/O rate.
The Node Resources portlet provides controls that let you choose what information is displayed. The information is refreshed every 60 seconds.
The PREFERENCES view allows you to choose which metrics appear in the portlet views.
About the Summary Views
The summary views provide a status of the resources on your Teradata Database system so you can monitor and locate issues. The NODES summary view displays the status of nodes, grouped by clique. The VPROCS summary view displays the status of vprocs in a node.
The following list describes the features in these views:
Selection Menus
Shows the system and resources currently being displayed. Choose a system, including multi-generational systems, and type of node.
Toolbar
Contains a count of the total resources on the selected system, the number of resources outside the threshold, and the number of down resources that are out of service or offline.
Teradata Viewpoint User Guide, Release 13.11
155
Chapter 17 Node Resources
Preview Sliders
Displays resources that are outside the threshold or down. You can customize the metrics displayed from the PREFERENCES view.
Resource Totals
Displays the total number of nodes or vprocs that are outside the threshold.
Clique ID
Displays the status of nodes, grouped by clique, when you select any of the Nodes views from the selection menus.
Node ID
Displays the status of vprocs in a node when you select any of the Vprocs views from the selection menus. You can click a node ID to view the details view for that node.
Preview Squares
Displays node or vproc status by color-coded squares, depending on the resource you chose from the selection menus.
Following is an example of the NODES summary view that you see when you select All
Nodes > Nodes > ALL from the selection menu.
156
Following is an example of the VPROCS summary view that you see when you select All
Nodes > Vprocs > ALL from the selection menu.
Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
Selecting a System to Monitor
You can use selection menus above the toolbar to choose a system and type of resource to monitor in the portlet, starting with the highest level menu on the far left. Different menu choices are available based on the previous menu.
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
6
7
8
9
Select a system from the menu.
Click Next.
Select a type of resource from the menu.
Click Next.
Select Nodes or Vprocs from the menu.
Click Next.
Do one of the following:
• Select Nodes, and then select ALL, With AMPs, or Without AMPs from the menu.
• Select Vprocs, and then select ALL, AMPs, or PEs from the menu.
[Optional] Click Next.
Selection Menu Choices
The Node Resources portlet displays data for a single Teradata Database system. The selection menu allows you to choose the type of data that appears in the summary views,
Teradata Viewpoint User Guide, Release 13.11
157
Chapter 17 Node Resources preview squares, and details views. Depending on your system, you may have access to multigenerational systems.
Selection Menu Choices Description
Nodes > ALL
Node data on all nodes, grouped by clique
Nodes > With AMPs
Node data on all nodes with AMPs, grouped by clique
Nodes > Without AMPs Node data on all nodes without AMPs, grouped by clique
Vprocs > ALL
Vprocs > AMP
Vproc data on all AMPs and vprocs, grouped by node
Vproc data only on AMPs, grouped by node
Vproc data only on PEs, grouped by node
Vprocs > PE
5500H, 5350, 4980, and so forth
Multi-generational system data
For example, by selecting 5500H, you can view and compare data for
5500H nodes within a system. Typically, you would not compare metrics on older hardware with metrics on newer hardware.
About the Toolbar
The toolbar allows you to filter the display attributes in the table view by using the resource count status buttons. The toolbar displays a summary count of:
• The total resources on the selected system
• The > threshold resources that are outside the threshold
• The number of down resources that are out of service or offline
After you click on any button, the table view appears with statistics and information about the selected resources to help you isolate issues.
158
About the Preview Sliders
Preview sliders display metrics of resources that are outside the threshold or down and help you determine if there is skew across the nodes or vprocs. The metrics and thresholds that appear in the preview sliders are set in the PREFERENCES view.
As you move a slider:
• A box to the right of each slider displays the total number of nodes or vprocs that are outside the threshold.
• A box above the slider displays a percentage + or - from the average. The left side of the slider represents the average. As you move the slider to the right, the metric is further away from the average.
• Preview squares reflect the resources that are outside the threshold.
You can click a node ID to view the details view for that node.
A single vproc might be counted for each resource total, however, that vproc is only counted once in the > threshold status button.
Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
About the Preview Squares
Preview squares are color-coded squares that help you determine status for a particular resource. Preview squares display data in the summary views and details views according to the resources selected from the selection menu.
Mouse over a preview square to display an information balloon containing the metrics for the resource selected. The information in the balloon corresponds to the metrics in the 3 preview sliders. Click a preview square to see the details view for that resource.
The following table describes the meaning of the preview squares.
Preview Square
Gray
Orange
Black
Status
Normal
Warning
Down
Description
No threshold is set or no metric is outside of the threshold.
At least one metric for a resource is outside the threshold.
Set automatically when the node goes down or when the
Teradata Database Administrator takes the resource out of service or offline.
Summary View Metrics
Select metrics to appear in the summary view for nodes, vprocs, AMPs, and PEs using the
Columns dialog box. Some metrics are only available in certain views.
N
Nodes > ALL, Nodes > With AMPs, Nodes > Without AMPs
V
Vprocs > ALL
A
Vprocs > AMP
Teradata Viewpoint User Guide, Release 13.11
159
Chapter 17 Node Resources
P
Vprocs > PE
Metric
CIC USAGE
CPU USAGE
DISK I/O
DISK OUT REQ
AVG
DISK READS
Description
Percent of Channel Interface Controller usage for this resource
Type
Percent
View
N
Percent of CPU usage that is spent being active (not idle) Percent
Number of disk I/Os for this resource Number
N, V, A, P
N, V, A
Number N, V, A Average number of outstanding disk requests for this resource
Total number of physical disk reads for this resource during the sample interval
Number N, V, A
Percent N
Number N, V, A
DISK USAGE Percentage of disk usage for this resource
DISK WRITES Total number of physical disk writes for this resource during the sample interval
HOST BLOCK
READS
HOST BLOCK
WRITES
Number of message blocks received from all hosts.
A message block may consist of multiple messages.
Number of message blocks sent to all hosts.
A message block may consist of multiple messages.
Number
Number
N, V, P
N, V, P
HOST I/O Number of host I/Os for this resource Number N, V, P
MEM AGINGS Memory agings and collections done on old segments Number N
Kilobytes allocated to memory resources Number N, V, A, P MEM
ALLOCATE KB
MEM
ALLOCATES
Segments allocated to memory resources Number N, V, A, P
MEM FAILURES Memory segment allocation attempts that failed Number N
NET A USAGE Total BYNET utilization (average of the online BYNETs) Percent N
Number N, V, A, P NET READS Messages read from the BYNET and input into the resource during the sample interval
NET WRITES Messages written from the resource and output to the
BYNET during the sample interval
Number N, V, A, P
Number N NV MEM
AGINGS
NV MEM
ALLOCATE
SEGS
NV MEM
ALLOCATE
Backup storage disk segments flushed from nonvolatile memory (on Teradata Database V2R6.x)
Complete and partial nonvolatile memory allocations of disk segments read (written) for backup storage (on
Teradata Database V2R6.x)
Kilobytes of nonvolatile memory allocated for AMP backup storage (on Teradata Database V2R6.x)
Number
Number
N, V, A
N
160 Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
Metric
PERCENT AMP
WORKER TASK
PERCENT
KERNEL
PERCENT
SERVICE
PERCENT
DISPATCHER
Description
CPU resources spent in AMP Worker Task (AWT) processing
Time spent in UNIX PDE Kernel waiting for I/Os to complete
CPU resources spent in UNIX PDE user service processing
Percent of CPU resources spent in PE Dispatcher processing
Type
Percent
Percent
Percent
Percent
View
V, A
N
N, V, A, P
V, P
PERCENT USER CPU resources spent in non-service user code processing Percent
PARSER USAGE Percent of CPU resources spent in PE Parser processing Percent
SWAPS Total number of swap reads and swap writes
SWAP DROPS Pages and segments dropped from resource memory during the sample period due to swapping
Number N
Number N
N
V, A, P
SWAP READS Segments read into resource memory from the disk during the sample interval due to swapping
SWAP WRITES Segments written to disk from resource memory during the sample interval due to swapping
Number N
Number N
About the Table View
The table view displays statistics and information about the resources at one time, helping you to isolate performance or system issues. This view is useful for comparing multiple vprocs. The table view can be accessed from the summary views by clicking a resource count status button in the toolbar or clicking any metric name in the preview slider.
Select and organize the displayed columns by clicking .
Click on a row to access the details view for a particular node or vproc, depending on the resource you chose from the selection menus.
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Teradata Viewpoint User Guide, Release 13.11
161
Chapter 17 Node Resources
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
\
*
?
Wildcard or
Symbol
=
>
>=
<
Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
*cat cat*l cat*
*ews*er
*%
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
Example Results
cat?
?cat
cats, catalog cat?l
cat???
p???er
scat, Scatter
catalog
catalog
packer, parser, proper
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
>60
>=60
<60
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
162 Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
!
Wildcard or
Symbol
<=
Description Example Results
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
<=60
!cat33
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Qualifying data is highlighted in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
Teradata Viewpoint User Guide, Release 13.11
163
Chapter 17 Node Resources
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
About the Details View
The details view displays statistics and information about a particular node or vproc to help you isolate performance or system issues. This view can be accessed by clicking on a preview square in the summary view or a row in the table view.
Preview squares at the bottom of the view help you determine status for a particular vproc within the node. Mouse over a square to display an information balloon containing metrics for that vproc. Mouse over other preview squares to compare the metric values with other vprocs. When viewing a vproc, the parent node number is shown in the information balloon.
Click on a preview square to directly access the details view for another vproc. Click on the node ID to view the details view for that node.
164
Details View Metrics
The following table lists the metrics that may appear in the details view.
Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
If the processing is skewed towards one or more AMPs (that is, the parallelism is not 100%), the metrics may display a value greater than 100%. This is due to the way the data is normalized.
Some metrics are only available in certain views.
N
Nodes > ALL, Nodes > With AMPs, Nodes > Without AMPs
V
Vprocs > ALL, Vprocs > AMP, Vprocs > PE
Metric/Name Description
STATUS Status of the node or vproc, where U means the resource is up, D means the resource is down, and S means the resource is in stand-by state
CLIQUE # Location of the clique in the rack. When there is one clique, a dash displays
NODE #
NODE TYPE
Unique identifier for each node (cabinet ID and node
ID)
Type of node
TYPE Type of vproc
AMP COUNT Number of AMPs currently executing session requests on this node
PE COUNT
HOST ID/
CLUSTER
Number of PEs currently active on this node
Host Id is the logical host identifier associated with a
PE or session. Cluster is the cluster number associated with an AMP.
Sessions currently logged on to this PE SESSION LOG
COUNT
SESSION RUN
COUNT
CPU USAGE
Sessions sending TSR messages to this vproc
PERCENT
KERNEL
% SERVICE
% USER
Percent of CPU usage that is not spent being active
(not idle)
Time spent in UNIX PDE Kernel waiting for I/Os to complete
Percent of CPU resources spent in UNIX-PDE user service processing
CPU resources spent in non-service user code processing
Type
Character
Number
Number
Various
Character
Number
Number
Number
Number
Number
Percent
Percent
Percent
Percent
PERCENT AMP
WORKER TASK
CPU Resources spent in AMP Worker Task (AWT) processing
Percent
PARSER USAGE Percent of CPU resources spent in PE Parser processing Percent
N
V
N
N
V
View
N, V
N
V
V
V
N, V
N
N, V
N
V
V
Teradata Viewpoint User Guide, Release 13.11
165
Chapter 17 Node Resources
Metric/Name Description
PERCENT
DISPATCHER
Percent of CPU resources spent in PE Dispatcher processing
NET A USAGE Total BYNET utilization (average of the online
BYNETs)
DISK USAGE Percent of disk usage for this resource
CIC USAGE Percent of Channel Interface Controller usage for this resource
DISK READS Total physical disk reads for this resource during the sample interval
DISK WRITES Total physical disk writes for this resource during the sample interval
DISK I/O
DISK OUT REQ
AVG
HOST BLOCK
READS
HOST BLOCK
WRITES
Number of disk I/Os
Average number of outstanding disk requests for this resource
Message blocks received from all hosts. A message block may consist of multiple messages.
Message blocks sent to all hosts. A message block may consist of multiple messages.
HOST I/O Number of host I/Os
SWAP READS Segments read into node memory from the disk during the sample interval due to swapping
SWAP WRITES Segments written to disk from node memory during the sample interval due to swapping
SWAP DROPS Pages and segments dropped from node memory during the sample period due to swapping
SWAPS
MEM
ALLOCATES or
MEM ALLOCS
Total number of swap reads and swap writes
Segments allocated to memory resources
Type
Percent
Percent
Percent
Percent
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
Number
MEM
ALLOCATE KB
MEM
FAILURES
Kilobytes allocated to memory resources
Segment allocation attempts that failed
Number
Number
MEM SEGS
AGINGS
NET READS
Memory agings and collections done on old segments Number
Messages read from the BYNET and input into the node during the sample interval
NET WRITES Messages written from the node and output to the
BYNET during the sample interval
Number
Number
N
N
N
N, V
N, V
N, V
N, V
N
View
V
N
N, V
N, V
N, V
N, V
N, V
N, V
N, V
N
N
N, V
N, V
166 Teradata Viewpoint User Guide, Release 13.11
Chapter 17 Node Resources
Metric/Name Description
NV MEM
AGINGS
Backup storage disk segments flushed from nonvolatile memory (on Teradata Database V2R6.x)
NV MEM
ALLOCATE
Kilobytes of nonvolatile memory allocated for AMP backup storage (on Teradata Database V2R6.x)
NV MEM
ALLOCATE
SEGS
PE CPU USE
Complete and partial nonvolatile memory allocations of disk segments read (written) for backup storage (on
Teradata Database V2R6.x)
CPU usage of the parsing engines
AVAILABLE
AWT
Total available AMP worker tasks
AWT IN USE Total in-use AMP worker tasks
Messages waiting on the vproc MESSAGE
COUNT
MESSAGE DQ
COUNT
Messages processed from the queue
Type
Number
Number
Number
Percent
Number
Number
Number
Number
About the Preferences View
The PREFERENCES view allows you to customize the way resource information is displayed in the portlet views.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
Slider Metrics
Select the metrics to display in the preview sliders for each view. Set the maximum threshold percent for each metric.
Display
Display or hide nodes or cliques in the summary views that are operating within set thresholds.
Setting Metric Thresholds
Use the Slider Metrics tab in the PREFERENCES view to set metric thresholds. Use
MaxThreshold % to set the maximum range of the preview sliders in the summary views.
1
2
3
4
5
From the portlet frame, click to access the PREFERENCES view.
Click the Slider Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a view from the list.
Select a metric from the list.
V
V
V
V
V
View
N
N
N, V
Teradata Viewpoint User Guide, Release 13.11
167
Chapter 17 Node Resources
6
7
8
Enter the maximum threshold percent for each metric.
The maximum range of the preview sliders changes in the summary views.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Hiding Nodes and Cliques
Use the Display tab in the PREFERENCES view to display or hide nodes or cliques in the summary views.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Display tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do one of the following:
• Select the Hide nodes or cliques that are operating within set thresholds check box to hide nodes or cliques in the summary views.
• Clear the Hide nodes or cliques that are operating within set thresholds check box to display nodes or cliques in the summary views.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
168 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 18
Productivity
Productivity
The Productivity portlet allows you to analyze performance trends for a system over a 24hour or a 48-hour period. You can also compare current performance with average performance measured across one or more weeks of historical data. Additionally, you can monitor and compare the productivity of multiple Teradata Database systems by adding a
Productivity portlet to the view for each system to be monitored.
Only the Teradata Viewpoint Administrator can enable system health metrics, configure the warning and critical thresholds, and customize the states and tooltips for your system. The metrics and thresholds are usually selected to highlight an unusual load on the system that has the potential to impact overall performance.
About the Productivity View
The PRODUCTIVITY view displays a high-level overview of the performance trends for a
Teradata Database system during the preceding 24 hours or 48 hours. Data trends are represented in a horizontal graph called a sparkline. Mouse over a sparkline to see an information balloon containing detailed information about the data point. There are three types of sparkline:
CANARY RESPONSE TIME
Canary-response-time metrics show the time (in milliseconds) a query takes to complete. Canary response times, such as SYSTEM HEARTBEAT are represented by a curve sparkline. Every 15 minutes, a new data point is added to the sparkline, representing the average canary response time recorded during the interval. A large dot and a number at the end of the sparkline indicate the last data point captured.
QUERIES PER HOUR
Queries-per-hour (QPH) metrics show the number of queries against the database for each hour over the 24-hour or 48-hour period for the selected metric. For example, the TOTAL QPH sparkline is a skyline sparkline, used to represent QPH metrics. A flat, solid data point represents the number of queries for each hour on the timeline.
SYSTEM HEALTH
The SYSTEM HEALTH metric provides an overview of the worst state of the monitored system sampled at 15-minute intervals during the 24-hour or 48-hour period. SYSTEM HEALTH is represented by a bar sparkline and uses default or
Teradata Viewpoint User Guide, Release 13.11
169
Chapter 18 Productivity custom states and tooltips. The values on the sparkline represent the worst system state experienced during each 15-minute period.
Productivity Metrics
You can select and organize metrics to display, set thresholds, adjust the vertical axis range for each metric, and adjust the duration over which the averages are calculated using the
PREFERENCES view.
Metric
Canary Response Times
Metrics
System Metrics: System
Health
Default
Sparkline
Curve
Bar
System Metrics: System
Heartbeat
System Metrics: Total
QPH
Curve
Skyline
Queries Per Hour Metrics Skyline
Past
Average
Line
Yes
No
Yes
Yes
Yes
Default
No
Yes
Yes
Yes
No
Description
Select individual canary query metrics
Shows customized states or these default states: healthy, warning, critical, down, or unknown.
Shows whether the Teradata Database system is responsive
Select total queries per hour metrics
Select individual queries per hour metrics
About the Preferences View
The PREFERENCES view allows you to customize the portlet to monitor key metric trends that are important to efficient operation of your Teradata Database system. You can select only one system to monitor for each portlet instance.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
System
Select a Teradata Database system to monitor.
170 Teradata Viewpoint User Guide, Release 13.11
Chapter 18 Productivity
Metrics
Select and organize metrics to monitor in the summary view. Drag metrics from a list of all available metrics to the Preview pane where you can preview the results as you go.
Settings
Define settings for the metrics selected for display. Set thresholds to highlight metric values that are outside normal operating ranges. Set maximum vertical-axis ranges to limit the range of values displayed for selected metrics.
Past Averages
Adjust the number of weeks of data points used to calculate the average value for each displayed metric.
Selecting a System to Monitor
Use the System tab in the PREFERENCES view to select a Teradata Database system to monitor. Only one system can be selected for each portlet instance.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the System tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a system from the list.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Selecting Metrics to Monitor
Use the Metrics tab in the PREFERENCES view to select and organize the metrics shown in the summary view.
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
The Preview pane uses sample data to show how metric rows are displayed in the summary view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do any of the following to change the way metric rows are displayed:
• Add a metric row. Drag a metric from the Select metrics for display list to the
Preview pane.
• Remove a metric row. Drag a metric from the Preview pane to the Metric Removal pane.
Teradata Viewpoint User Guide, Release 13.11
171
Chapter 18 Productivity
5
6
• Change metric row order. Drag a metric already in the Preview pane to a new location in the pane (up or down).
Changes to the metrics affect only the system currently selected, provided that the metrics are available on that system.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Metric Thresholds
Use the Settings tab in the PREFERENCES view to set the THRESHOLD and VERTICAL
AXIS RANGE values for the metrics selected for display. Only metrics selected using the
Metrics tab appear in the list of available metrics.
Only the Teradata Viewpoint Administrator can enable System Health metrics and configure the degraded and critical thresholds. These selections are made in the Teradata
Systems portlet.
1
2
3
4
5
6
7
From the portlet frame, click to access the PREFERENCES view.
Click the Settings tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
[Optional] Enter the THRESHOLD setting for each available metric.
The threshold line does not appear in the view when the value entered is zero or blank.
[Optional] Enter the VERTICAL AXIS RANGE for each available metric.
The sparkline scales automatically when the value entered is zero or blank.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Past Averages
Use the Past Averages tab in the PREFERENCES view to specify the number of weeks of data points used to calculate the average values displayed in metric graphs.
1
2
3
4
5
From the portlet frame, click to access the PREFERENCES view.
Click the Past Averages tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter the number of weeks (1 to 99 weeks).
The default is 2 weeks.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
172 Teradata Viewpoint User Guide, Release 13.11
6
Click OK.
Chapter 18 Productivity
Teradata Viewpoint User Guide, Release 13.11
173
Chapter 18 Productivity
174 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 19
Query Monitor
Query Monitor
The Query Monitor portlet allows you to view information about queries running in a
Teradata Database system so you can spot problem queries. You can analyze and decide whether a query is important, useful, and well written. After you have identified a problem query, you can take action to correct the problem by changing the priority or workload, releasing the query, or aborting the query or session. You can take these actions for one query or session, or multiple queries or sessions at a time.
The summary view contains a table with one row allocated to each of the sessions, account strings, users, or utilities running on the database.
The portlet allows you to filter queries in all of the session views. You can set thresholds for any column and when the threshold is exceeded, the information is highlighted in the sessions table.
Select a row to access session and query information in the details view.
Using Query Monitor, you can also determine the types of utilities that are running most frequently on the system and then set utility limits. You can spot utilities that are using a large number of partition connections and, potentially, a high number of resources.
From the PREFERENCES view, you can set the criteria values used to display sessions in the My Criteria view and customize the information displayed in the views. Set criteria values to display only those sessions currently running on the selected system that exceed the specified criteria. For example, you can troubleshoot Teradata Database system problems to quickly explore details about queries such as the current state of a query or how long a query has been blocked.
Monitor multiple Teradata Database systems by opening additional instances of the Query
Monitor portlet.
About the Query Monitor View
The QUERY MONITOR view displays a summary of sessions, account strings, users, or utilities running on the selected Teradata Database system.
Selection Menu
Shows the system and resources currently being displayed. Choose a system and view options including session, account string, user, or utility.
Teradata Viewpoint User Guide, Release 13.11
175
Chapter 19 Query Monitor
Toolbar
Provides a count of the sessions in each of the states. Click on a state to remove all states except the selected state from the view.
Filters
Filters allow you to change displayed data by showing only rows that match your filter criteria. You can also sort on the column headers to find the details you want.
Summary Table
Provides summary information about sessions, account strings, users, or utilities in columns. The current view is configured in the Columns dialog box and
PREFERENCES view and the settings that include the most data are used. The view is refreshed every 30 seconds. Click on a row in the table to see details.
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Abort allows you to abort the selected query or session.
Change Workload allows you to change the workload of the selected query or session.
Release Query allows you to release the selected query from a delay queue.
Export creates a .csv file containing the report information as shown.
176
Selecting a System to Monitor
You can use selection menus above the toolbar to choose a system, type of view, and information to monitor in the portlet, starting with the highest level menu on the far left.
Different menu choices are available based on the previous menu.
You are not required to make a selection from every menu each time you modify a selection.
Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
1
2
3
4
5
6
7
In the selection menu, click the currently selected system name to display a list of available systems.
Select a system from the menu.
Click Next.
Select a type of view from the menu.
Click Next.
Select an additional type of view from the menu.
Click Next.
Selection Menu Choices
The Query Monitor portlet displays data for a single Teradata Database system. The selection menu defines the information and metrics that appear in the summary view, session views, details views, and utility view.
Selection Menu Choices Description
By Session > ALL
Displays the sessions table for all sessions
By Session > My Criteria Displays the sessions table for sessions that exceed the set criteria values and the metrics enabled in the PREFERENCES view
By Account String
By User
By Utility
Displays rows of account strings in the sessions table
Displays rows of users in the sessions table
Displays rows of utilities in the sessions table
About the Bar Graph
The bar graph provides a Top sessions by menu where you can select one of several metrics.
A few seconds later, the graph refreshes to display sessions with the largest value compared with the total number of sessions currently running. Click on any session ID in the bar graph to see session details.
Use the maximized view to view some of the smaller sessions by clicking Other.
The Display tab in the PREFERENCES view allows you to show or hide the bar graph in the summary view.
About the Toolbar
The toolbar allows you to specify the display attributes of the view.
The query count status buttons provide near-real-time status of the number of sessions in each state category. Click on any query count status button to filter for the selected category.
Teradata Viewpoint User Guide, Release 13.11
177
Chapter 19 Query Monitor
For example, click DELAY to display the sessions running a query that is queued waiting to run in the selected Teradata Database system.
ALL
Sessions currently running in the selected Teradata Database system.
NOT IDLE
Sessions in any state except idle.
ACTIVE
Sessions running a query that is in progress.
BLOCK
Sessions running a query that is waiting for a locked resource such as a database table or view.
DELAY
Sessions running a query that is queued waiting to run.
ABORT
Sessions running a query that is in the process of aborting (rolling back changes made by the query).
RESP
Sessions running a query that has completed and is sending (responding) spooled data back to the user.
IDLE
Sessions not currently running a query.
PARSE
Sessions running a query that is being parsed. It has not begun to execute.
OTHER
Sessions whose status is unknown and do not fall into any of the above categories.
QTDELAYED
Sessions delayed due to a queue table restriction.
SESDELAYED
Utility sessions that are on the workload delay queue.
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the
178 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
\
*
?
Wildcard or
Symbol
=
>
>=
Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
*cat cat*l cat*
*ews*er
*%
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
cat?
?cat
cat?l
cat???
p???er
Example Results
cats, catalog
scat, Scatter
catalog
catalog
packer, parser, proper
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
>60
>=60
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
Teradata Viewpoint User Guide, Release 13.11
179
Chapter 19 Query Monitor
!
Wildcard or
Symbol
<
<=
Description
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
Example Results
<60 59, 58, 50, 8, and so forth
<=60
!cat33
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Values exceeding the threshold are highlighted and displayed in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
180 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Symbol Reference
The following symbols appear in the portlets:
Symbol Name Definition
ABORTING Query has been aborted, and changes are being rolled back.
ACTIVE
BLOCKED
Query is running.
Query is waiting for a lock held by another query.
DELAYED
IDLE
OTHER
PARSING
Query is in a delay queue waiting to run.
No query is running.
Query is in an unknown state.
Query is being parsed before running.
QTDELAYED Query is waiting for rows to be inserted into a queue table.
RESPONSE Query is returning results to the user.
SESDELAYED Query is delayed because a utility limit has been exceeded in FastLoad,
MultiLoad, FastExport, or ARC.
Summary View Metrics
Metrics appearing in the summary view can be selected using the Columns dialog box.
Metrics available for monitoring and display are listed below.
Metric
ACCOUNT
BLOCKED TIME
CPU SKEW
CPU USE
Description
Account from which a query was submitted
How long the query has been blocked
Type
Character
Number
CPU skew during the last sample Percent
Percent of the total amount of available CPU used by the query Percent
Teradata Viewpoint User Guide, Release 13.11
181
Chapter 19 Query Monitor
Metric
Delta CPU
Delta I/O
DURATION
HOST
IMPACT CPU
Description
Total CPU usage time consumed, in seconds, since the last sample
I/O count since the last sample
How long the query has been running
Host ID or LAN ID associated with the PE that processed the logon request for the session
CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
How long the query has been in the current state
I/O skew during the last sample
Partition in which the query is running
IN STATE
I/O SKEW
PARTITION
PJI
QUERY BAND
REQ CPU
REQ I/O
SESSION ID
SPOOL
START
STATE ICON
STATE
SYSTEM
Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
Entire query band string (query bands are a set of name-value pairs defined by the user to tag sessions or transactions with an ID through an SQL interface)
Number of CPU seconds needed to run the query
Number of disk I/Os performed to run the query
Unique session identifier
Amount of spool space the query requires
Time that the query started running on Teradata Database
Icon representing the current state of the query
Text describing the current state of the query
Full name of the system running the query
TEMP SPACE Amount of temp space used by the query
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
USERNAME
WORKLOAD
Name of the user who submitted the query
Workload in which the query is running
Type
Number
Number
Number
Number
Number
Number
Percent
Character
Number
Character
Number
Number
Number
Number
Number
Icon
Character
Character
Number
Number
Character
Character
User and Account String Metrics
Metrics appearing in the user or account string views can be selected using the Columns dialog box. Metrics available for monitoring and display are listed below.
182 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
Metric
ACCOUNT
USERNAME
Description
Account from which a query was submitted
Name of the user who submitted the query
Type
Character
Character
TOTAL
ACTIVE COUNT
ABORTING
BLOCKED
IDLE
PARSING
RESPONSE
DELAYED
Total number of sessions for the user or account (total of
Active, Aborting, Blocked, Idle, Parsing, Response, Delayed, and Unknown)
Number of queries that are running
Queries that have been aborted and changes are being rolled back
Queries waiting for a lock held by another query.
Queries that are not running
Queries being parsed before running
Queries that are returning results to the user
Queries in a delay queue waiting to run
Number
Number
Number
Number
Number
Number
Number
Number
UNKNOWN COUNT Number of queries in an unknown state
CPU USE
Number
Percent of the total amount of available CPU used by the query Percent
DELTA I/O
SPOOL
Number
Number
DELTA CPU
IMPACT CPU
Delta REQUEST
COUNT
I/O count since the last sample
Current spool used by the current request across all AMPs, in bytes
Total CPU usage time consumed, in seconds, since the last sample
Total CPU seconds consumed across all AMPs that are active for this session
Requests initiated by the session
Number
Number
Number
About the Utility View
The utility view displays utility names, states, and limits in the following columns:
Name
The type of utility. The name also includes the partition on which the utility is being run. Click a name to obtain details about an individual utility.
Note: TDWM Statistics reports many active utilities other than utility session, so if you select an active utility to obtain more information, you might receive a
No
Data
message.
Active
The current number of utilities active on the partition.
Teradata Viewpoint User Guide, Release 13.11
183
Chapter 19 Query Monitor
Limit
The maximum number of utilities allowed to be active on the partition. The limit value can be either the system default value for the utility or a rule throttle value.
You can override the default limits by changing the setting in the DBS control record.
About the Details View
The details view displays statistics and information about the selected session. This view can be accessed by clicking on a session row in the summary view.
When viewing a request, you can see detailed read-only information from the following tabs:
Overview
Key statistics for a session. Any value exceeding preset thresholds is highlighted.
SQL
SQL for the selected query.
Explain
Explain steps for the query, including step statistics and Explain text.
Blocked By
Details about other queries that are blocking this query.
Query Band
Displays the query band name and value for the selected query.
Delay
Details about rules delaying this query.
Use to change the priority or workload, release a query, or abort a query or session for one query or session at a time.
Use the and buttons to move through sessions without returning to the summary view.
184 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
Details View Metrics
Query Information
Metric
STATE
CPU USE
CPU
CPU Δ
Description
Query state, such as ACTIVE, BLOCKED, TERMINATE
Type
Character
Number TIME IN STATE
TOTAL DURATION
I/O
How long the query has been in the current state, displayed as
HH:MM:SS
How long the query has run, displayed as HH:MM:SS
Percent of the total amount of available CPU used by the query Percent
Total CPU usage time, in seconds, consumed by the query Number
Total CPU usage time consumed, in seconds, since the last sample
Number of I/O operations the query performed
Number
Number
Number
I/O Δ
CLASSIFICATION
MODE
I/O count since the last sample
Indicator that a query or session matches a workload definition (WD) if Teradata DWM Category 3 is enabled.
Available values are:
Number
Character
CPU SKEW
AUTO = Workload was assigned automatically by TASM
REQUEST = Workload was assigned manually at the
Request Level
SESSION = Workload was assigned manually at the
Session Level
This field is empty if this is not a DBC/SQL session or if
Teradata DWM Category 3 is disabled.
CPU skew during the last sample Percent
Teradata Viewpoint User Guide, Release 13.11
185
Chapter 19 Query Monitor
Metric
I/O SKEW
SPOOL
Description
I/O skew during the last sample
Amount of spool space the query is using
TEMP SPACE
WORKLOAD
Amount of temp space the query is using
Workload where the query is actively running
PJI Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
IMPACT CPU CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
Type
Percent
Number
Number
Character
Number
Number
Number
Session Information
Metric
USER
ACCOUNT
SOURCE
PARTITION
REQUESTS
Description
Name of the user that submitted the query
Account of the user that submitted the query
Source details, such as application name, IP address, and host user name
Partition in which the query is running
Number of queries submitted by the session
Type
Character
Character
Character
Character
Number
About the Overview Tab
The Overview tab provides detailed information about key metrics for the selected session and its queries. The metric values displayed provide a snapshot of Teradata Database system status. Metrics that exceed defined thresholds are highlighted. Under SESSION INFO, you can click on a user or account to see all sessions for that user or account string.
About the SQL Tab
The SQL tab displays the SQL for the selected query. This information is read-only. Use to export the SQL as a .txt file.
About the Explain Tab
The Explain tab displays an abbreviated version of the Step statistics and the Explain text that result from an Explain request in an SQL session. Each Explain step is uniquely identified with a number where the background of the number box indicates status. This information is read-only. This tab only appears when a query meets certain thresholds established by the Teradata Viewpoint Administrator. Use to export the Explain data as a .csv file.
186 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
Step Information
Step Number
Description
• Completed steps are at the top of the list and indicated by a black number box.
• Active steps are indicated by a pulsating number box (flashes gray and white).
• Steps to run are at the bottom of the list and indicated by a white number box.
Confidence Level Indicator Icon
EST. TIME
EST. ROWS
ACTUAL TIME
ACTUAL ROWS
- No confidence in the estimate
- Low confidence in the estimate
- High confidence in the estimate
- High confidence in the estimate due to a join index
Estimated execution time for the step.
Estimated number of rows for the step.
Actual CPU time consumed by the step, or blank if the step has not run.
Actual number of rows for the step, or blank if the step has not run.
About the Blocked By Tab
The Blocked By tab displays details about other queries that are blocking this query. This information is read-only. This tab is only available when the selected query is blocked. The
Session ID is a link that allows you to drill down into the session information for the blocking session. This allows you to navigate the chain of blocking sessions to identify the session currently holding the lock on the object.
Blocked By Information
USERNAME
HOST
SESSION ID
LOCK TYPE
STATUS
LOCKED
Description
Name of the user that is running the query that holds the lock.
Host ID or LAN ID associated with the PE that processed the logon request for the session.
Session ID of the blocking query.
Type of lock. Type can be Exclusive, Read, Write, or Access.
Lock status. Status can be Waiting or Granted.
Name of the locked object.
About the Query Band Tab
The Query Band tab displays the session and transaction query bands for the selected query.
This information is read-only.
Query Band Information
NAME
Description
The name of the query band.
Teradata Viewpoint User Guide, Release 13.11
187
Chapter 19 Query Monitor
Query Band Information
VALUE
Description
The value of the query band.
About the Delay Tab
The Delay tab displays details about rules that are delaying this query. The Delay tab is available if the system being monitored is a Teradata Database 13.10 system or newer. The information on the Delay tab is read-only, and no actions are available from this tab. This tab is available when the selected query is delayed. The Delay tab displays all rules that apply to the query and a scroll bar appears if there are more than two rules.
Delay Information
BLOCKING COUNT
Description
Number of consecutive times this session has blocked at least one other session.
Amount of time request has been delayed by a utility throttle rule.
UTILITY THROTTLE DELAY
TIME
SYSTEM THROTTLE DELAY
TIME
WORKLOAD THROTTLE
DELAY TIME
RULE NAME
RULE TYPE
OVERRIDABLE
Amount of time request has been delayed by a system throttle rule.
Amount of time request has been delayed by a workload throttle rule.
Name of rule causing request delay.
TASM type of rule causing request delay.
Indicates if the Teradata Database Administrator can abort or release the request.
About Managing Queries and Sessions
You can manage queries and sessions to improve workload performance.
Abort
Abort the selected query or session.
Change Priority
Change the priority of the selected query or session.
Change Workload
Change the workload of the selected query or session.
Release Query
Release the selected query from a delay queue.
You must log on with a user ID that has permissions to abort, change priorities or workloads, or release queries. If you log out, close a portlet, or open a new portlet instance, you must log on again.
Note: Change Workload is available only if Teradata Active System Management (TASM) is enabled. If TASM is disabled, Change Priority is available. If you do not see Change
188 Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
Workload or Change Priority in the menu, the system you are monitoring does not support these features or you do not have permission to use them. If the query you are monitoring is delayed, only Release Query is available.
Aborting a Query or a Session
You can abort a query or a session that is blocking other queries or consuming too many resources.
1
2
3
4
5
6
7
8
From the summary view, click the row of the query you want to abort.
Click .
Select Abort.
[Optional] Log on.
Click Next.
Select one of the following:
• Abort Query to abort the selected query.
• Abort Session to abort the selected query and log off the session.
Click Next.
Click Next to confirm your selection.
Changing the Priority of a Query or Session
You can change the priority of a query or session to allow higher priority queries to run or to balance session resources.
This menu item is only available when workloads are not enabled and the system being monitored supports this feature.
1
2
3
From the summary view, click the row of the query with a priority that needs to change.
Click .
Select Change Priority.
4
5
6
7
8
9
10
[Optional] Log on.
Click Next.
From the list of accounts that have been assigned to the user, select an account to copy the account name to the Edit new account string box.
[Optional] Edit the account string or enter a new account string.
Click Next.
Select one of the following:
•
Move just this query to change the priority of the selected query.
• Move this query and all future queries associated with this session to change the priority of the selected query and all subsequent queries in the current session.
Click Next.
Teradata Viewpoint User Guide, Release 13.11
189
Chapter 19 Query Monitor
190
11
Click Next to confirm your selection.
Changing the Workload of a Query or Session
You can change the workload of a query or session to either allow higher priority workloads to run or to balance workload resources.
This menu item is only available when workloads are enabled and the system being monitored supports this feature.
1
2
3
4
5
6
7
8
9
10
From the summary view, click the row of the query with the workload that needs to change.
Click .
Select Change Workload.
[Optional] Log on.
Click Next.
[Optional] Select a different workload from the list.
Click Next to confirm your selection.
Select one of the following:
• Move just this query to change the workload of the selected query.
•
Move this query and all future queries associated with this session to change the workload of the selected query and all subsequent queries in the same session.
Click Next.
Click Next to confirm your selection.
Releasing a Query
You can release a query that is in the delay queue and waiting to run.
1
2
3
4
5
6
From the summary view, click the row of the query you want to release.
Click .
Select Release Query.
[Optional] Log on.
Click Next.
Click Next to confirm your selection.
Changing Multiple Sessions
You can change system resources for multiple sessions, users, or account strings in the following locations:
• All sessions table from the summary view
• My Criteria sessions table from the summary view
• User sessions table from the details view
• Account strings table from the details view
• Utility table from the details view
Teradata Viewpoint User Guide, Release 13.11
Chapter 19 Query Monitor
1
2
3
4
5
6
7
Select By Session in the selection menu and one of the following choices:
• All
• My Criteria
[Optional] Access user or account string tables from the details view.
a
Click on a session.
b
Click the USER or ACCOUNT link under SESSION INFO.
Select a session action from the Display Configuration menu.
Check boxes appear next to the sessions.
Do one of the following:
• Select the check box in the column heading to select all sessions.
• Select the check boxes for specific sessions.
Select Next.
[Optional] Log on.
Follow the instructions on subsequent screens.
About the Preferences View
The PREFERENCES view allows you to customize the way query and session information is displayed in the portlet views.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
Display
Choose formatted or unformatted SQL to display in the SQL tab. Display or hide the bar graph in the portlet.
Criteria
Set the criteria values used to display sessions in the My Criteria view. After setting the criteria value and enabling a metric, any query exceeding the criteria value appears in the sessions table in the My Criteria view.
Selecting Display Options
Use the Display tab in the PREFERENCES view to select a format for the SQL and to display or hide the Top Sessions graph.
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Click the Display tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do one of the following:
Teradata Viewpoint User Guide, Release 13.11
191
Chapter 19 Query Monitor
5
6
7
• Select the Format SQL check box to use formatted SQL.
• Clear the Format SQL check box to use unformatted SQL.
Do one of the following:
• Select the Display Top Sessions graph check box to display the Top Sessions bar graph in the portlet. remove the top sessions graph from the portlet.
• Clear the Display Top Sessions graph check box to remove the Top Sessions bar graph from the portlet.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Selecting Threshold Criteria
Use the Criteria tab in the PREFERENCES view to set the criteria values used to display sessions in the My Criteria view. If any enabled metric exceeds the threshold value, the session is displayed when the metric is also selected in the Columns dialog box.
1
2
3
4
5
6
7
From the portlet frame, click to access the PREFERENCES view.
Click the Criteria tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter threshold values for each metric you want tested.
Select the metrics you want enabled.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
192 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 20
Query Spotlight
Query Spotlight
The Query Spotlight portlet allows you to view completed queries that exceeded the thresholds you set. You can view individual queries of interest as a line graph over time and against a variety of metrics. The Snapshot tab displays the query data at any point in time.
Use Query Spotlight to:
• View general information about a query.
• Inspect a query as a line graph using the date-range slider and time marker.
• Set a date and time in the rewind toolbar. This feature is useful for rewinding data in surrounding portlets to a point of time specified within a query. The Query Spotlight portlet can set time and date but it does not respond to changes made in the rewind toolbar.
The PREFERENCES view allows you to choose the query criteria that filter which metrics appear in the views.
You can monitor multiple systems by opening additional instances of the Query Spotlight portlet. Rewind control is given over to the last instance of Query Spotlight to use the time marker.
About the Query Spotlight View
The QUERY SPOTLIGHT view displays completed queries on the selected system and time frame.
Selection Menu
Choose the system and time frame to display. Time frame options include viewing by hours, days, weeks, or setting a custom range.
Filters
Filters allow you to change displayed data by showing only rows that match your filter criteria. You can also sort on the column headers to find the queries you want.
Queries
Provides summary information about each query. The current view is configured in the Columns dialog box and PREFERENCES view and the settings that include the most data are used. Click on a row in the table to see details.
Teradata Viewpoint User Guide, Release 13.11
193
Chapter 20 Query Spotlight
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Export creates a .csv file containing the query information as shown.
194
Selecting a System to Monitor
You can use the selection menu to choose a system and time frame to monitor.
1
2
3
4
5
In the selection menu, click the currently selected system name to display a list of available systems.
Select a system from the menu.
Click Next.
Select a time frame.
Hours, week, and month are all based on the current time. Custom is a fixed time period.
Click Next.
If you chose Custom, enter a date and time range and click OK.
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Teradata Viewpoint User Guide, Release 13.11
Chapter 20 Query Spotlight
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
\
*
?
Wildcard or
Symbol
=
>
>=
<
Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
*cat cat*l cat*
*ews*er
*%
=CAT
=cat?
=Cat*
=Cat_
=Cat\?
*\?
=cat\* cat\?
Example Results
cat?
?cat
cats, catalog cat?l
cat???
p???er
scat, Scatter
catalog
catalog
packer, parser, proper
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
CAT
cats
Cat, Cats, Catalog
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
>60
>=60
<60
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
Teradata Viewpoint User Guide, Release 13.11
195
Chapter 20 Query Spotlight
!
Wildcard or
Symbol
<=
Description
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
Example Results
<=60
!cat33
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Values exceeding the threshold are highlighted and displayed in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
196 Teradata Viewpoint User Guide, Release 13.11
Chapter 20 Query Spotlight
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Query Spotlight View Metrics
Query Information
Metrics appearing in the queries table can be selected using the Columns dialog box.
Metrics available for monitoring and display are listed below.
Metric
CPU SKEW
DURATION
END
HOST
IMPACT CPU
I/O SKEW
PJI
Description
CPU skew during the last sample
Type
Percent
How long the query has been running Number
Time that the query stopped running on Teradata Database Number
Number Host ID or LAN ID associated with the PE that processed the logon request for the session
CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
Number
I/O skew during the last sample
Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
Percent
Number
REQ CPU
REQ I/O
REQ NUMBER
SESSION ID
Number of CPU seconds needed to run the query
Number of disk I/Os performed to run the query
Number of the specific request
Unique session identifier
SPOOL
START
Amount of spool space the query requires
Time that the query started running on Teradata Database
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
Number
Number
Number
Number
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
197
Chapter 20 Query Spotlight
Metric
USERNAME
Description
Name of the user who submitted the query
Type
Character
About the Details View
The details view displays a graph, statistics and information about a selected query. Choose a metric from the metrics list to display one query as a line graph.
Metrics List
List of metrics that can be graphed.
Date-Range Slider
The date-range slider appears for longer queries to control the amount of data displayed in the graph.
The darker-shaded area between the slider handles represents the current date range being displayed in the graph.
Time Marker
Pinpoints information about a query at a particular point in time and is used to rewind time in the rewind toolbar when Control Rewind is enabled.
Previous/Next Query
Use the and buttons to move through the queries.
Time Frame
Displays the date and time range duration of the query.
Job Steps
The line graph contains vertical dashed lines denoting the boundaries between
Explain steps. Job step numbers represent the Explain step that occurs between the dashed lines and are shown just below the graph area. Overlapping steps or stepswithin-steps are not shown.
When there is not enough room to display individual steps, the numbers are grouped together. For example, you might see 1, 2, 3, 4, 8. Step 8 represents steps 5,
6, 7 and 8.
You can see detailed read-only query information from the following tabs:
Overview
Key statistics for a session. Any value exceeding set thresholds is highlighted.
SQL
SQL for the selected query.
Explain
Explain steps for the query, including step statistics and Explain text.
198 Teradata Viewpoint User Guide, Release 13.11
Chapter 20 Query Spotlight
Blocked By
Details about other queries that are blocking this query.
Query Band
Displays the query band name and value for the selected query.
This tab appears only when a query band is included in the query.
Snapshot
Key statistics for the query. Any value exceeding set thresholds is highlighted.
Adjusting the Date Range
The date-range slider shows the most recent part of the query. The date and time range are displayed in the time frame above the slider on the right.
1
Mouse over the date-range slider to activate the start-time and end-time handles.
Teradata Viewpoint User Guide, Release 13.11
199
Chapter 20 Query Spotlight
200
2
3
Drag the start-time handle until the intended start time is displayed in the time frame.
Drag the end-time handle until the intended end time is displayed in the time frame.
4
[Optional] Drag the date-range handle left or right to move the start-time and end-times handles in unison.
Selecting a Metric to Graph
You can select different metrics to graph. The metrics contain historical details you can use to investigate queries.
1
2
Click next to the metric name.
Select a metric from the menu.
About Query Spotlight and Rewind
The Query Spotlight portlet cannot rewind and does not use the rewind toolbar. The
Control Rewind button in the Query Spotlight portlet is a toggle that enables portal rewind for portlets that support the rewind function on the same page. When rewind is enabled, the time marker in Query Spotlight sets the time and date in the rewind toolbar and controls the time and date for data in the other portlets. This allows you to investigate how metrics compare across portlets during the same time frame.
When Control Rewind is toggled off, the time marker no longer controls the date and time for data in the other portlets and portal rewind is controlled from the rewind toolbar.
Enabling and Using Rewind Control
The time marker is a vertical line in the graph with small triangles at the top and bottom. Its position represents a point in time and controls rewind for the portlets that support the rewind function. The time marker appears green when Control Rewind is enabled, and you interact with it.
1
2
3
From the Details View, click Control Rewind located below the graph in the middle of the page.
The icon in the Control Rewind button turns green and the rewind toolbar appears.
Click on or drag the time marker to a specific point in time.
[Optional] Drag the time marker to another point in time.
Disabling Rewind Control
The icon in the Control Rewind button is green and the rewind toolbar is displayed with rewind is enabled.
Teradata Viewpoint User Guide, Release 13.11
Chapter 20 Query Spotlight
1
Click Control Rewind.
The in the Control Rewind button turns gray. The time marker no longer controls the time and date in the rewind toolbar, and portal rewind is controlled from the rewind toolbar.
About the Overview Tab
The Overview tab provides detailed information about key metrics for the selected query.
Metrics that exceed defined thresholds are highlighted.
Metric Description
TOTAL DURATION How long the query has run, displayed as HH:MM:SS
USER Name of the user that submitted the query
ACCOUNT
REQ NUMBER
REQ CPU
REQ I/O
REQ MAX SPOOL
SOURCE
Account of the user that submitted the query
Number of I/O operations the query performed
Total CPU usage time, in seconds, consumed by the query
Number of I/O operations the query performed
Amount of spool space the query requires
Source details, such as application name, IP address, and host user name
Type
Number
Character
Character
Number
Number
Number
Number
Character
About the SQL Tab
The SQL tab displays the SQL for the selected query. This information is read-only. Use to export the SQL as a .txt file.
About the Explain Tab
The Explain tab displays an abbreviated version of the Step statistics and the Explain text that result from an Explain request in an SQL session. Each Explain step is uniquely identified with a number where the background of the number box indicates status. This information is read-only. This tab only appears when a query meets certain thresholds established by the Teradata Viewpoint Administrator. Use to export the Explain data as a .csv file.
Step Information
Step Number
Description
• Completed steps are at the top of the list and indicated by a black number box.
• Active steps are indicated by a pulsating number box (flashes gray and white).
• Steps to run are at the bottom of the list and indicated by a white number box.
Confidence Level Indicator Icon
- No confidence in the estimate
- Low confidence in the estimate
Teradata Viewpoint User Guide, Release 13.11
201
Chapter 20 Query Spotlight
Step Information
EST. TIME
EST. ROWS
ACTUAL TIME
ACTUAL ROWS
Description
- High confidence in the estimate
- High confidence in the estimate due to a join index
Estimated execution time for the step.
Estimated number of rows for the step.
Actual CPU time consumed by the step, or blank if the step has not run.
Actual number of rows for the step, or blank if the step has not run.
About the Blocked By Tab
The Blocked By tab displays details about other queries that are blocking this query. This information is read-only. This tab is only available when the selected query is blocked.
Blocked By Information
USERNAME
HOST
SESSION ID
LOCK TYPE
STATUS
LOCKED
Description
Name of the user that is running the query that holds the lock.
Host ID or LAN ID associated with the PE that processed the logon request for the session.
Session ID of the blocking query.
Type of lock. Type can be Exclusive, Read, Write, or Access.
Lock status. Status can be Waiting or Granted.
Name of the locked object.
About the Query Band Tab
The Query Band tab displays the session and transaction query bands for the selected query.
This information is read-only.
Query Band Information
NAME
VALUE
Description
The name of the query band.
The value of the query band.
About the Snapshot Tab
The Snapshot tab displays detailed query data for any point in time. Metrics that exceed set thresholds are highlighted.
Snapshot Information
Metric
STATE
TIME IN STATE
CPU USE
Description
Query state, such as ACTIVE, BLOCKED, TERMINATE
Type
Character
How long the query has been in the current state, displayed as
HH:MM:SS
Number
Percent of the total amount of available CPU used by the query Percent
202 Teradata Viewpoint User Guide, Release 13.11
Chapter 20 Query Spotlight
Metric
CPU
CPU Δ
I/O
I/O Δ
CPU SKEW
I/O SKEW
SPOOL
TEMP SPACE
WORKLOAD
CLASSIFICATION
MODE
Description
Total CPU usage time, in seconds, consumed by the query
Total CPU usage time consumed, in seconds, since the last sample
Number of I/O operations the query performed
I/O count since the last sample
CPU skew during the last sample
I/O skew during the last sample
Amount of spool space the query is using
Amount of temp space the query is using
Workload where the query is actively running
Indicator that a query or session matches a workload definition (WD) if Teradata DWM Category 3 is enabled.
Available values are:
AUTO = Workload was assigned automatically by TASM
REQUEST = Workload was assigned manually at the
Request Level
SESSION = Workload was assigned manually at the
Session Level
PJI Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
IMPACT CPU
This field is empty if this is not a DBC/SQL session or if
Teradata DWM Category 3 is disabled.
CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
Type
Number
Number
Number
Number
Percent
Percent
Number
Number
Character
Character
Number
Number
Number
Time Marker Information
Metric
TIME LOCATION
STEP LOCATION
Description
Explain step in which the time marker is located
Type
Temporal location of the time marker within the Explain step Number
Number
About the Preferences View
The PREFERENCES view allows you to set thresholds to filter query data and customize the way query list information is displayed in the portlet view.
Teradata Viewpoint User Guide, Release 13.11
203
Chapter 20 Query Spotlight
From the portlet frame, click to access the PREFERENCES view and the Criteria tab.
Criteria
Specify threshold criteria to limit the number of queries shown in the query table.
After enabling a metric filter and setting the criteria value, any query exceeding the criteria value appears in the query table if it is also selected in the Columns dialog box.
Selecting Threshold Criteria
Use the Criteria tab in the PREFERENCES view to set the criteria values used to display queries in the QUERY SPOTLIGHT view. If any enabled metric exceeds the threshold value, the query is displayed when the metric is also selected in the Columns dialog box.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter threshold values for each metric you want tested.
Select the metrics you want enabled.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
204 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 21
Remote Console
Remote Console
The Remote Console portlet allows you to run many of the Teradata Database console utilities remotely from within the Teradata Viewpoint portal.
Using this portlet, you can:
• Select or search for a system.
• Select or search for a utility.
• Enter console utility commands.
• Display responses from the commands.
Teradata field engineers, Teradata Database operators, System Administrators, and System
Programmers use Teradata Database utilities to administer, configure, monitor, and diagnose issues with Teradata Database.
About the Remote Console View
The REMOTE CONSOLE view allows you to select and run Teradata Database console utilities. You can run only one utility at a time for each portlet instance. Permission to use each console utility is granted to individual users based on their role and portlet access. If a message appears in the console denying access to a utility, you are not authorized to use that utility.
The following list describes the features in this view:
Selection Menus
Select a system and utility.
Clear Console
Clear the content from the console.
Disconnect
Disconnect the utility from Teradata Database.
Console
Display responses from the commands.
Input Line
Enter commands.
Teradata Viewpoint User Guide, Release 13.11
205
Chapter 21 Remote Console
Submit Command
Click to submit the command.
206
Running a Utility
You can select a different system and utility, enter commands, and use the keyboard up and down arrow keys to access previous commands.
Use this procedure only if you are already logged into a system and utility, and want to select a different system or utility.
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
6
7
Select a system from the menu.
Click Next.
Select a utility from the menu
Click Next.
The name of the utility appears in the console.
Enter a command in the input line.
Click to submit the command.
The command results appear in the console.
Teradata Viewpoint User Guide, Release 13.11
Chapter 21 Remote Console
8
[Optional] From the input line, use your keyboard up and down arrows to access previous commands.
These commands are temporarily saved and are available as long as you are running the utility. After disconnecting from the utility, these commands are cleared.
Note: After approximately 30 minutes of inactivity in the portlet, the Teradata Database connection times out, and you must reconnect to the utility and Teradata Database.
About the Remote Console Menus and Toolbar
The menus and toolbar allow you to specify the display attributes of the REMOTE
CONSOLE view.
Use the selection menus to:
• Select or search for a system.
• Select or search for a utility.
Use Clear Console to clear the content from the console.
Use Disconnect to disconnect from the utility and Teradata Database.
Selecting a System and Utility
You can use selection menus above the toolbar to choose a system and utility to monitor in the portlet, starting with the highest level menu on the far left. You must select a system and utility before you can submit commands and see the output in the console.
1
2
3
4
5
In the selection menu, click the currently selected system name to display a list of available systems.
Enter the name of the system in the Search box or select a system from the menu.
The Search box is not case sensitive. If you cannot find the correct system, press the
Backspace key on your keyboard until you find the system you want in the list.
As you type, the list shows the systems that match your filter criteria.
Click Next.
Enter the name of the utility in the Search box or select a utility from the menu.
The Search box is not case sensitive. If you cannot find the correct utility, press the
Backspace key on your keyboard until you find the utility you want in the list.
As you type, the list shows the utilities that match your filter criteria.
Click Next.
Clearing the Console
1
From the summary view, click Clear Console.
The contents of the console no longer appear.
Teradata Viewpoint User Guide, Release 13.11
207
Chapter 21 Remote Console
Disconnecting from a Utility
You can disconnect from a utility and database.
1
From the summary view, click Disconnect.
The input line no longer accepts input and the portlet no longer responds to commands.
The Teradata Database connection closes when you remove the portlet from the portal page or click Disconnect. If you close the browser without clicking Disconnect, the
Teradata Database connection times out in approximately 30 minutes.
About Console Utilities
Teradata field engineers, Teradata Database operators, System Administrators, and System
Programmers use Teradata Database utilities to administer, configure, monitor, and diagnose issues with Teradata Database.
Console Utilities
Information about available console utilities is provided in the following table. You can access only the following subset of Teradata Database utilities.
Utility
Abort Host
Check Table
Configure
DBS Control
Ferret
Description
Cancel all outstanding transactions running on a failed host until the system restarts the host.
Check for inconsistencies between internal data structures, such as table headers, row identifiers, and secondary indexes.
Define AMPs, PEs, and hosts, and describe their interrelationships for Teradata
Database.
Display and modify the DBS Control Record fields.
Define the scope of an action, such as a range of tables or selected vprocs, display the parameters and scope of the action, and perform the action. The action is to either move data to reconfigure data blocks and cylinders, or display disk space and cylinder free space percent in use.
Gateway Global Monitor and control the sessions of Teradata Database LAN-connected users.
Lock Display Display a snapshot capture of all real-time database locks and their associated currently running sessions.
Operator
Console
Priority
Scheduler
Run supervisor commands to manage the programs that perform Teradata
Database operations.
Create, modify, and monitor Teradata Database process prioritization parameters.
All processes have an assigned priority based on their Teradata Database session.
Priority Scheduler allocates CPU and I/O resources based on the priority.
Query
Configuration
Display the current Teradata Database configuration, including the node, AMP, and PE identification and status.
Query Session Monitor the state of selected Teradata Database sessions on selected logical host
IDs.
208 Teradata Viewpoint User Guide, Release 13.11
Chapter 21 Remote Console
Utility
Recovery
Manager
Show Locks
Description
Display information used to monitor progress of a Teradata Database recovery.
Display locks placed by archive and recovery, and by table rebuild operations on databases and tables.
Teradata DWM
Dump
Display information about active Teradata Dynamic Workload Manager
(Teradata DWM) rules on a Teradata Database system.
Vproc Manager Manage the vprocs including obtaining the status of specified vprocs, initializing vprocs, forcing a vproc to restart, and forcing a Teradata Database restart.
Information on how to use the majority of these utilities is located in Utilities. Information on how to use Teradata DWM Dump is in Teradata Dynamic Workload Manager User
Guide. Information on how to use Operator Console is in Graphical User Interfaces:
Database Window and Teradata MultiTool. Experienced utilities users can refer to the simplified command descriptions in Utilities Quick Reference which provides the syntax diagrams for each Teradata Database utility.
You can access these documents from Teradata Information Products at: http:// www.info.teradata.com
.
About the Preferences View
The PREFERENCES view allows you to change the console text and background colors.
From the portlet frame, click to access the PREFERENCES view.
Setting the Display Background
Use the Display tab in the PREFERENCES view to change the console text and the display background to light or dark.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Display tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Change the color of the text and background by selecting one of the following options:
•
Light text on dark background
•
Dark text on light background
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
209
Chapter 21 Remote Console
210 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 22
Space Usage
Space Usage
The Space Usage portlet allows you to monitor Teradata Database disk space usage and reallocate permanent disk space from one database to another. Select a system and one of the following reports:
Database: By most space
Displays the total, currently in use, and greatest amount in use of permanent disk space for a database.
Vproc: By space
Displays the permanent disk space by processor.
The summary view allows you to monitor the database perm space, temp space, and spool space. The details view allows you to view perm space usage over time. You can reallocate permanent disk space from either of these views.
The PREFERENCES view allows you to customize the way information is displayed in the summary and details view and set thresholds for filters.
About the Space Usage View
The SPACE USAGE view displays a summary of perm, spool, and temp space for the
Teradata Databases running on the selected system. Use this information to see if you need to reallocate disk space to maximize space usage.
The following list describes the features in this view:
Selection Menus
Shows the system and space usage report. You can choose a different system and report.
Toolbar
Shows the number of databases that exceed a set threshold of perm space, and the number of users that exceed a set percentage threshold of spool or temp space.
Filters
Click any space usage threshold button to filter for the selected category. Set the thresholds in the PREFERENCES view.
Changes the displayed data to only show rows that match your filter criteria.
Teradata Viewpoint User Guide, Release 13.11
211
Chapter 22 Space Usage
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Export creates a .csv file containing the report information as shown.
Add Space
Reallocates permanent disk space from one database to another.
Report
Shows space usage information.
About the Space Usage Menus and Toolbar
The selection menus and toolbar allow you to specify the display attributes of the SPACE
USAGE view.
212
Use the selection menus to select a system and report. You can save these selections for subsequent sessions by selecting the Use these settings as defaults check box in the
Teradata Viewpoint User Guide, Release 13.11
Chapter 22 Space Usage
PREFERENCES view. If you have not selected this check box, you see the first system you have permission to view in the Database: By most space report.
The toolbar displays the number of databases over a set threshold. View how many users are over a set percentage threshold of spool or temp space. Click any space usage threshold button to filter for the selected category. Define the thresholds in the PREFERENCES view.
Selecting a System and Report
You can use selection menus above the toolbar to choose a system and report, starting with the highest level menu on the far left.
1
In the selection menu, click the currently selected system name to display a list of available systems.
2
3
4
5
Select a system from the menu.
Click Next.
Select a report from the SELECT REPORT dialog box.
Click Next.
The report appears in the summary view.
Searching for Available Disk Space
Search for available disk space to determine which databases have the least amount of disk space so you can reallocate space if necessary.
1
2
3
From the selection menus, select Database: By most space.
Click Next.
The report appears in the summary view.
Sort on a column by clicking the column header.
Adding Space
You can reallocate permanent disk space from one database to another. Permission to reallocate disk space is granted to individual users based on their role and portlet access.
1
To reallocate space, do one of the following:
• From the summary view, click located next to the database name, and then select
Add Space from the menu that appears.
2
3
• From the details view, click Add Space.
Log on with a user ID that has permission to add space.
Click Next.
Teradata Viewpoint User Guide, Release 13.11
213
Chapter 22 Space Usage
4
5
6
7
In the Add text box, enter how much space you want to reallocate.
In the from text box, select the source database from where the perm space is being taken.
Click Submit.
A confirmation message appears.
Click OK.
The space is reallocated to the database.
About Filters and Sorting
Filters allow you to change displayed data by showing only rows that match your filter criteria. Use > or < filter symbols for columns with decimals because numeric metric values are rounded in the display. You can narrow the search further by filtering on multiple columns.
Sorting allows you to change displayed order on all pages when there are multiple pages of table data. Sort on a column by clicking the column header. A second click sorts in descending order. You can sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
The filtering, sorting, and page number settings that you choose for the default or minimized view are not used when you switch to the maximized view.
Filter Symbols
Use a wildcard character or symbol in the filter to search for words that have spelling variations or contain a specific pattern of characters. Some symbols can be used to search for numbers.
*
?
Wildcard or
Symbol
=
Description
Use this wildcard character to match alpha or numeric characters in the position it occupies. Type this wildcard character at the beginning, middle, or end of your search.
This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this wildcard character to match zero, one, or multiple alpha or numeric characters in the position it occupies.
Type this symbol at the beginning, middle, or end of your search. This wildcard character can be used more than once in the same search.
This wildcard character can be used in conjunction with any other symbol.
Use this symbol at the beginning of your search to match alpha or numeric characters literally. The search results are case-sensitive.
This symbol can be used in conjunction with * and ?.
Example Results
cat?
?cat
cats, catalog cat?l
cat???
p???er
scat, Scatter
catalog
catalog
packer, parser, proper
*cat cat*l cat*
*ews*er
*%
=CAT
=cat?
=Cat*
cat, cats, catalog, scatter, wildcat
catalog
cat, cats, catalog newscaster, newspaper what is 100%?
CAT
cats
Cat, Cats, Catalog
214 Teradata Viewpoint User Guide, Release 13.11
Chapter 22 Space Usage
!
\
Wildcard or
Symbol
>
>=
<
<=
Description
Use this symbol in front of a wildcard character so the wildcard is interpreted as a regular character and not as a wildcard.
This symbol can be used in conjunction with *, ?, and \.
Use this symbol to match any number that is greater than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only > and the number.
Use these symbols in conjunction to match values greater than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only >= and the number.
Use this symbol to match any number that is less than the specified value. This symbol is only used to perform mathematical comparisons.
For columns containing percentages, avoid using % in the filter. Type only < and the number.
Use these symbols in conjunction to match values less than or equal to the specified number.
For columns containing percentages, avoid using % in the filter. Type only <= and the number.
Use this symbol at the beginning of your search to match alpha or numeric characters that do not contain the alpha or numeric characters.
This symbol can be used in conjunction with any other symbol or wildcard character.
=Cat_
=Cat\?
*\?
=cat\* cat\?
>60
>=60
<60
<=60
!cat33
Example Results
Cat_
Cat?
what is 100%?
cat*
cat?, Cat?
61, 62, 70, 500, and so forth
60, 61, 62, 70, 500, and so forth
59, 58, 50, 8, and so forth
60, 59, 58, 50, 8, and so forth
cat32, cat34, and so forth
Clearing Filters
You can clear the filter box filters from the table.
1
Do one of the following:
• To clear individual filter boxes, click .
• To clear all filter boxes, click and choose Clear Filters.
Configuring Columns to Display
Use the Columns dialog box to set thresholds and select, lock, and order columns. You can resize columns in the table.
Teradata Viewpoint User Guide, Release 13.11
215
Chapter 22 Space Usage
216
1
2
3
4
5
6
7
8
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click Set, type a threshold value, and click OK.
Values exceeding the threshold are highlighted and displayed in the table.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] In the table, drag the column heading border in either direction to resize the column.
The thresholds and column selection, order, and lock settings that you choose for the default or minimized view are not saved when you switch to the maximized view.
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
Space-Reporting Metrics
The metrics (reports) displayed in the summary view are selected using the selection menus.
Descriptions for the metrics available for monitoring and display are listed according to their report name:
•
Database: By most space
•
Vproc: By space
Database: By most space
This report shows the total, currently in use, and greatest amount in use of permanent disk space for the database.
Column
Name
Description
Database name
Type
Alpha
Teradata Viewpoint User Guide, Release 13.11
Chapter 22 Space Usage
Column
Max Perm
Current Perm
Current Perm %
Peak Perm
Peak Perm %
Unused Perm %
Current Perm Skew
%
Description
Total amount of permanent disk space available for the database
Type
Number
Amount of total permanent disk space the database is currently using Number
Percent Percentage of the total permanent disk space the database is currently using (Current Perm divided by Max Perm)
Greatest amount of permanent disk space the database has used
Highest percentage of the total permanent disk space the database has used (Peak Perm divided by Max Perm)
Percentage of the total permanent disk space currently not being used (100% minus Current Perm %)
A measure of perm space data distributed across the AMPs for the selected database
Note: A perfectly distributed database (same amount of data on each
AMP) has a skew of 0. A database which has all the data on one AMP has a skew of 100.
Number
Percent
Percent
Percent
Max Spool
Current Spool
Current Spool %
Peak Spool
Peak Spool %
Max Temp
Current Temp
Total amount of temporary spool space available for the database Number
Amount of total temporary spool space the database is currently using Number
Percentage of the total temporary spool space the database is currently using (Current Spool divided by Max Spool)
Percent
Greatest amount of temporary spool space the database has used
Highest percentage of the total temporary spool space the database has used (Peak Spool divided by Max Spool)
Total amount of temporary space available for the database
Amount of total temporary space the database is currently using
Current Temp % Percentage of the total temporary space the database is currently using (Current Spool divided by Max Spool)
Peak Temp Greatest amount of temporary space the database has used
Peak Temp %
Parent Database
Highest percentage of the total temporary space the database has used (Peak Temp divided by Max Temp)
Parent database name
Number
Percent
Number
Number
Percent
Number
Percent
Alpha
Vproc: By space
This report shows the permanent disk space, by processor.
Column
Vproc ID
Max Perm
Description
AMP number
Total amount of permanent disk space available for the database, by processor
Type
Number
Number
Teradata Viewpoint User Guide, Release 13.11
217
Chapter 22 Space Usage
Column
% in Use
Description
Current Perm Amount of total permanent disk space the database is currently using, by processor
Percentage of the total permanent disk space the database is currently using, by processor (Current Perm divided by Max Perm)
Available Perm Amount of total permanent disk space currently not being used by the database, by processor (Max Perm divided by Current Perm)
Type
Number
Percent
Number
About the Space Usage Details View
The details view displays perm space usage information for tables. Click the database name in the summary view to access the details view.
Use the sparkline to determine if the perm space usage has increased or decreased over time.
The following list describes the features in this view:
Add Space
Reallocate permanent disk space from one database to another.
Display Configuration
Clear Filters removes any filters you have set in the filters boxes.
Configure columns allows you to choose columns to display and set thresholds.
Export creates a .csv file containing the report information as shown.
Sparkline
See the perm space usage over the last 3 months in the sparkline, where each data point shown on the sparkline is the amount of perm space usage recorded during a single day. Mouse over the sparkline to display detailed information about the perm space usage in an information balloon.
Filters
View the peak and max perm space usage listed to the right of the sparkline.
Change displayed data by removing rows that do not match your filter criteria.
Report
View perm space usage information for tables.
Sorting
Sort on a column by clicking the column header.
Columns
Select and organize the displayed columns by clicking .
218 Teradata Viewpoint User Guide, Release 13.11
Chapter 22 Space Usage
About the Preferences View
From the portlet frame, click to access the PREFERENCES view. Set the threshold values for the perm, spool, and temp space buttons in the toolbar for the Database: By most
space report. Customize the way information is displayed in the summary and details views.
Setting Filter Thresholds and Display Options
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Click the General tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter specific threshold values for buttons in the summary view toolbar for the
Database: By most space report.
5
6
7
Select the Define KB, MB and GB as multiples of 1,000 instead of 1,024 check box to change the display in the summary and details views.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
219
Chapter 22 Space Usage
220 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 23
SQL Scratchpad
SQL Scratchpad
The SQL Scratchpad portlet allows you to enter SQL queries and retrieve query results from a Teradata Database system. You can connect to a system to run a single- or multistatement query, view and export the query results, and save one or more queries and pin, or temporarily save, results. With the Object Browser, you can view a list of database objects and insert object names into queries.
About the Query Box
The query box allows you to build queries and submit them to a Teradata Database system.
The results are available below the query box after you run the queries.
System Name
Connects to a Teradata Database system. Displays name of currently connected system, or allows you to connect to a different system.
Disconnect
Disconnects from the Teradata Database system displayed in the System Name.
Query Box
Allows you to enter and edit SQL queries.
Load
Shows the query history from the current session or saved queries from any session.
The list displays the OBJECT BROWSER.
Run
Runs statements in the query box and provides results in the query tabs. The list allows you to run a selected query and to save all queries or selected queries.
Teradata Viewpoint User Guide, Release 13.11
221
Chapter 23 SQL Scratchpad
222
Connecting to a System
You can connect to a Teradata Database system. If you are currently connected to a system, you can switch to another system by clicking System Name.
1
2
3
4
5
6
Click Select System.
The SELECT SYSTEM dialog box appears.
Select or search for a system from the list.
Teradata Database systems appear for those systems you are authorized to access.
Click Next.
In the CONNECT TO SYSTEM dialog box, enter the following information:
a b
User Name
Password
c
[Optional] Account String
[Optional] Select a character set from the list.
UTF8 is the default character set.
Click Connect.
Disconnecting from a System
You can disconnect from a Teradata Database system to avoid idle connections.
1
Click Disconnect.
Select System replaces System Name in the portlet frame. The results table is cleared.
Running a Query
You can enter and send multiple queries to a Teradata Database system. The currently connected system is System Name in the portlet frame.
1
2
Enter one or more queries into the query box.
Click Run to submit all the queries you entered.
Teradata Viewpoint User Guide, Release 13.11
Chapter 23 SQL Scratchpad
Running a Selected Query
You can run one or more adjacent queries if you do not want to run all queries that you typed in the query box.
1
2
In the query box, highlight one or more complete query statements you want to run.
When you select more than one adjacent query statement to run, highlight each statement, including the final semicolon.
Click the list next to Run and select Run Selected Query.
About Query Statements
You can enter single- or multi-statement SQL queries using Data Definition Language
(DDL), Data Control Language (DCL), or Data Manipulation Language (DML) statements in the query box.
DDL statements
Begin with SQL keywords such as CREATE, ALTER, DROP, DELETE, MODIFY,
HELP, SHOW, SET, REPLACE.
DCL statements
Begin with SQL keywords GRANT, REVOKE, GIVE.
DML statements
Begin with SQL keywords such as SELECT, INSERT, UPDATE.
For single-statement queries, enter the query, optionally followed by a semicolon. For multistatement queries, each statement must end with a semicolon.
Cancelling a Long-Running Query
Occasionally, a query may take longer to execute than you expected. You can cancel the query before it completes.
1
To stop a single, long-running query, click Cancel in the results table.
Saving All Queries
You can save all queries that you enter in the query box for reuse in current or future sessions.
1
2
3
4
Click the list next to Run and select Save All Queries.
In the ENTER QUERY NAME dialog box, enter a name for your set of queries.
Click Save.
[Optional] To view your set of saved queries, click Load and click the Saved tab.
Saving Selected Query
You can save one or more adjacent queries that you typed in the query box for reuse in current or future sessions.
Teradata Viewpoint User Guide, Release 13.11
223
Chapter 23 SQL Scratchpad
1
2
3
4
5
In the query box, highlight one or more complete query statements you want to save.
When you select more than one adjacent query statement to save, highlight each statement, including the final semicolon.
Click the list next to Run and select Save Selected Query.
In the ENTER QUERY NAME dialog box, enter a name for the selection.
Click Save.
[Optional] To view your saved query selection, click Load and click the Saved tab.
About Query Results
Query results are displayed in a table below the query box. You can view, pin, or export query results, as well as view SQL statements, Explain steps, and blocked query information.
Query Tabs
Display detailed query results in separately numbered tabs. Pinned query tabs show stored results temporarily for the current session.
Pins or unpins query results to be stored in the current session.
Submitted Statement
Displays a portion of the SQL statement.
Results Tab
Displays a table of query results that can be exported.
SQL Tab
Displays the entire SQL statement for the selected query.
Explain Tab
Provides Explain steps for the query, including step statistics and explain text. Only appears when a query meets certain thresholds established by the Teradata
Viewpoint Administrator.
Blocked By Tab
Provides details about other queries that are blocking this query. Only appears when a query is blocked.
Export
Saves results of the currently selected Query tab to a comma-separated values (CSV) file.
224 Teradata Viewpoint User Guide, Release 13.11
Chapter 23 SQL Scratchpad
Results for each statement in the query appear in a separately numbered Query tab. Results for a single-statement query appear in the Query 1 tab. Results for multi-statement queries appear in additional tabs. For example, if you submit three statements, results appear in the
Query 1, Query 2, and Query 3 tabs.
Viewing Query Results
You can view results for queries that you submit in the Query tabs.
1
Click the Query tab corresponding to your query listed in the query box.
A table with the results of your query appears in the Results tab.
Pinning Query Results
You can pin query results for later reuse in the session. If you do not pin the Query tab, new queries overwrite results currently displayed in the tab.
1
2
Click the Query tab that contains the query you want to pin and click .
The tab name changes from Query to Pinned. The tab retains its number. Query results remain pinned until you unpin the tab or disconnect from the Teradata Database system.
[Optional] To unpin your query, click .
The tab name changes from Pinned to Query. The results displayed in the results tab are cleared when you submit another query.
Exporting Table Data
You can export data displayed in portlet tables to a comma-separated values (CSV) file for further analysis and formatting. The exported CSV file contains data for all columns available in the portlet's table. The format for time, date, and some numeric values is different in the portlet's table than in the exported CSV file.
Teradata Viewpoint User Guide, Release 13.11
225
Chapter 23 SQL Scratchpad
226
1
2
3
Click in the table header and select Export.
Save the file using options in the browser's dialog box.
The file is saved to your download area or to a location that you specify, depending on the browser settings.
[Optional] Reformat exported data to match the format in the portlet table.
About the Results Tab
The Results tab provides detailed information about your query results. To view more columns in a wide results table, drag the scroll box to the right. This information is read-only.
About the SQL Tab
The SQL tab displays the SQL for the selected query. This information is read-only.
About the Explain Tab
The Explain tab displays an abbreviated version of the Step statistics and the Explain text that result from an Explain request in an SQL session. Each Explain step is uniquely identified with a number where the background of the number box indicates status. This information is read-only. This tab only appears when a query meets certain thresholds established by the Teradata Viewpoint Administrator.
Step Information
Step Number
Description
• Completed steps are at the top of the list and indicated by a black number box.
• Active steps are indicated by a pulsating number box (flashes gray and white).
• Steps to run are at the bottom of the list and indicated by a white number box.
EST. TIME
EST. ROWS
ACTUAL TIME
ACTUAL ROWS
Estimated execution time for the step.
Estimated number of rows for the step.
Actual CPU time consumed by the step, or blank if the step has not run.
Actual number of rows for the step, or blank if the step has not run.
About the Blocked By Tab
The Blocked By tab displays details about other queries that are blocking this query. This information is read-only, and no actions are available from this tab. This tab is available as long as other queries continue to block this query. When the blocking query conditions are resolved, this query can then complete, and the tab is no longer displayed.
Blocked By Information
USERNAME
SESSION ID
LOCK TYPE
Description
Name of the user that is running the query that holds the lock.
Session ID of the blocking query.
Type of lock. Type can be Exclusive, Read, Write, or Access.
Teradata Viewpoint User Guide, Release 13.11
Chapter 23 SQL Scratchpad
Blocked By Information
LOCKED
Description
Name of the locked object.
About the Object Browser
The OBJECT BROWSER allows you to view a list of objects in a Teradata Database system and insert an object into a query. You can reduce the time required to build a query and reduce errors when inserting object names in a query.
The Object Browser displays database objects in the connected system as a hierarchically organized tree. Types of database objects include databases, users, tables, views, macros, functions, and procedures. You can use filtering to search for object types in the tree.
• Expand a branch of objects in the tree by clicking next to the object type. The Object
Browser totals the objects found on each expanded branch by object type. If no objects of the type are found, the branch displays 0 items.
• Refresh the list of objects in the tree by clicking next to the Teradata Database system name.
About Filtering Objects
Filtering allows you to display only objects that match your filter criteria in the OBJECT
BROWSER. Filtering is not case sensitive.
Filter boxes are located at each branch and sub-branch of the directory tree, under the name of the database object. When you click the object name, the filter box appears.
• Filter for databases and users by entering text in the box under the top branch of the tree.
For example, if you type dbc
in the filter box at the top branch of the tree under the
Teradata Database name, databases and users containing the string DBC are displayed in the tree.
• Filter for all other database object types by entering text in the box under the object type.
Object types include Tables, Views, Macros, Functions, and Procedures.
For example, if you type clear
in the filter box under a Macros sub-branch of the tree, macros containing the string Clear are displayed in the tree.
Change the filter criteria by entering a new string in the filter box. Unless you type new filter criteria or open other branches, the Object Browser maintains the state of the tree, including opened branches and applied filter criteria.
You cannot use
?
or
*
as wildcards to match single or multiple characters in filter criteria.
The Object Browser can display a maximum of 500 objects at each branch of the tree. To display fewer results, enter filter criteria to narrow the list of database objects displayed on the branch.
Object Types
This table lists database objects and associated icons displayed in the OBJECT BROWSER
Teradata Viewpoint User Guide, Release 13.11
227
Chapter 23 SQL Scratchpad
Database Object
Database
User
Table
View
Macro
Function
Procedure
Icon
Inserting an Object into a Query
You can insert the qualified name of a database object into a query in the query box using the
OBJECT BROWSER. A qualified object name is a name composed of more than one identifier, joined with the dot character (.). The identifiers together specify a database object with more precision than a single name.
1
2
3
4
5
6
7
8
9
Position the cursor in the query box where you want to insert the object.
Click the list next to the Load button and select Insert Object.
The OBJECT BROWSER opens, displaying the name of the connected Teradata
Database system.
Click next to the Teradata Database system name to expand the directory tree.
[Optional] Enter filter criteria in the box under the Teradata Database system to find databases or users containing the criteria.
Click next to the database object type in the tree to expand that sub-branch.
[Optional] Enter filter criteria in the box under the object type to find objects containing the criteria.
Click to highlight the object name.
Click Insert Object to copy the qualified object name to the query box.
The object name is inserted in the query box after the cursor location.
Click Close.
Example of Inserting an Object into a Query
If you select the table object Accounts of the DBC database in the Teradata Database system
DEV1, the selected identifier is inserted as a qualified object name into the query box.
1
2
3
Click DEV1 to open the directory tree.
Click DBC to open a database sub-branch of the directory tree. DBC is a database identifier for the DEV1 system.
Click Tables to open all table objects under the DBC sub-branch.
228 Teradata Viewpoint User Guide, Release 13.11
Chapter 23 SQL Scratchpad
4
5
Click Accounts in the sub-branch to select the table. Accounts is the identifier of a particular table in the DBC database for the DEV1 system.
Click Insert Object to copy the qualified name to the query box.
DBC.Accounts
appears after the cursor in the query box.
About the History Tab
The History tab provides the following information in the table or in mouseover text for queries submitted in the current session:
• Date and time of the query
• Complete SQL statement
• Amount of time to complete the query
• Number of rows processed
For longer SQL statements, click to display the complete query.
You can click Load to copy the query to the query box, for later reuse or editing.
Viewing Previously Run Queries
You can view a table of queries submitted in the current session. You can load a query from the table for reuse in the same session.
1
2
Click Load and click the History tab.
The History tab provides the following information in the table or in mouseover text on queries submitted in the current session:
• Date and time of the query
• Complete SQL statement
• Amount of time to complete the query
• Number of rows processed
Click Close.
Loading a Previously Run Query
You can copy a query submitted in the current session to the query box for reuse or editing.
1
2
3
4
Click Load and click the History tab.
In the query history table, click Load for the query you want to run.
The query is copied to the query box and the History tab closes.
[Optional] Edit the query before running it.
Click Close.
About the Saved Tab
The History tab provides the following information on queries saved in the current or previous sessions:
Teradata Viewpoint User Guide, Release 13.11
229
Chapter 23 SQL Scratchpad
• Name of the saved query
• Complete SQL statement
For longer SQL statements, click to display the complete query.
You can click Load to copy the query to the query box, for later reuse or editing. You can also click to delete a saved query.
Viewing Saved Queries
You can view a table of queries that you saved. You can also load a query from the table for reuse in the current or future sessions.
1
2
Click Load and click the Saved tab.
The name of the query and the SQL statement appear in the table of saved queries.
Click Close.
Loading a Saved Query
You can load a saved query into the query box for reuse or editing.
1
2
3
4
Click Load and click the Saved tab.
In the table of saved queries, click Load for the query you want to run.
The query is copied to the query box and the Saved tab closes.
[Optional] Edit the query in the query box before running it.
Click Close.
Deleting a Saved Query
You can delete a saved query that you no longer need.
1
Click Load and click the Saved tab.
2
3
4
In the table of saved queries, click for the query you want to delete.
A confirmation message appears.
Click OK.
Click Close.
230 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 24
System Health
System Health
The System Health portlet displays the status of monitored Teradata Database systems using a defined set of metrics and thresholds. This portlet reports status with customized states or these default states: healthy, warning, critical, down, or unknown with customized or default tooltips so you can investigate metrics exceeding healthy thresholds.
Only the Teradata Viewpoint Administrator can enable system health metrics, configure the warning and critical thresholds, and customize the states and tooltips for your system. The metrics and thresholds are usually selected to highlight an unusual load on the system that has the potential to impact overall performance.
About the System Health View
The SYSTEM HEALTH view displays a summary of overall states for one or more systems, using colored text and icons to represent system health. From the SYSTEM HEALTH summary view, click anywhere on a status icon or text to investigate metrics for that
Teradata Database system in the SYSTEM HEALTH DETAILS view.
Default Icon Text Color Default
States
Blue
Definition
Healthy All metrics are within healthy ranges.
Orange
Red
Warning At least one metric exceeded the warning threshold.
Critical At least one metric exceeded the critical threshold.
Black
Black
Down The selected system is down.
Unknown Status of the selected system is unknown.
Teradata Viewpoint User Guide, Release 13.11
231
Chapter 24 System Health
If your system does not use the default states and tooltips, your Teradata Viewpoint
Administrator customized the states and tooltips. You can customize the icons using the
PREFERENCES view, Icons tab.
About the System Health Details View
The SYSTEM HEALTH DETAILS view displays detailed statistics and information about each metric to evaluate the overall health of a Teradata Database system.
This view presents detailed metric information in several ways:
METRIC
Lists the names of the monitored metrics, using color to indicate metric health.
Displays metric descriptions when you mouse over the name.
VALUE VS THRESHOLD
Shows the current metric value in terms of warning and critical thresholds, indicated by gaps in the bar. Displays threshold values and default or custom tooltips when you mouse over the bar.
232
The at the end of some bar graphs indicates an unbounded metric, which have numerical values with no upper limit. The is red when the unbounded metric value exceeds 1.2 times the critical threshold.
LAST nn MIN
Displays a sparkline with the trend in metric values over the selected duration of 15,
30, or 60 minutes, suggesting whether the system is expected to degrade or improve in the future. Line color indicates metric health.
VALUE
Shows the metric value at the last Teradata Database system sampling, using color to indicate metric health.
Teradata Viewpoint User Guide, Release 13.11
Chapter 24 System Health
Viewing Component Down Details
When the COMPONENT DOWN metric has a value of one or greater, the SYSTEM
HEALTH DETAILS view displays detailed information so that you can investigate the extent of your exposure to components that are down.
1
2
From the SYSTEM HEALTH DETAILS view, click the COMPONENT DOWN metric name or value.
View the list of component types and the number of components affected.
System Health Details View Metrics
Metrics appearing in your SYSTEM HEALTH details view are selected for you by the
Teradata Viewpoint Administrator. Metrics available for selection and display are listed below.
Metric Description
CPU UTILIZATION Average node CPU use for User, System, and Wait IO
(CPU) USER
(CPU) SYSTEM
Average percentage of CPU time spent in user mode
Type
Percent
Percent
Average percentage of CPU time spent in system (kernel) mode Percent
(CPU) WAIT IO
NODE CPU SKEW
NODE IO SKEW
AMP CPU SKEW
Average percentage of CPU time spent waiting for I/O
Comparison of CPU use on the busiest node to the average node Percent
Comparison of disk use on the busiest node to the average node Percent
Comparison of CPU use on the busiest AMP to the average
AMP
Percent
Percent
AMP IO SKEW Comparison of disk use on the busiest AMP to the average AMP Percent
AMP WORKER TASKS Average number of AMP worker tasks in use on each AMP Number
TOTAL DISK SPACE Percentage of total disk space currently in use
MAX DISK SPACE BY
AMP
Available disk space currently in use
Percent
Percent
COMPONENT
DOWN
MEMORY USED
Number of components, such as BYNETs or AMPs that are down
Largest amount of memory used by a single node (KB)
Number
MAX SPOOL BY AMP Available spool space in use
MEMORY FAILURES Memory failures across all nodes
DBC DISK SPACE
ACTIVE SESSIONS
Available DBC disk space in use
Number of sessions with active queries
Number
Percent
Number
Percent
Number
Number SYSTEM HEARTBEAT Canary query showing the response time of the most recent system heartbeat (ms)
Additional Canaries Additional canary queries that are defined for the system and used as part of the monitoring of system health
Number
Teradata Viewpoint User Guide, Release 13.11
233
Chapter 24 System Health
About the Preferences View
The PREFERENCES view allows you to customize the portlet to display information that is most useful to you.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
Systems
Select Teradata Database systems to monitor in the SYSTEM HEALTH and
SYSTEM HEALTH DETAILS views.
Icons
Select an icon style to use in the SYSTEM HEALTH view.
Trend Interval
Select the interval for the SYSTEM HEALTH trend line. Choose from among 15, 30, or 60 minutes of historical data.
Selecting Systems to Monitor
Use the Systems tab in the PREFERENCES view to select Teradata Database systems to monitor in the summary and details views.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Systems tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select the check boxes of systems to display.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Selecting the System Health Icon Style
Use the Icons tab in the PREFERENCES view to select the icon style to display in the
SYSTEM HEALTH view.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Icons tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select the icon style to use in the SYSTEM HEALTH view.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
234 Teradata Viewpoint User Guide, Release 13.11
Chapter 24 System Health
Setting the System Health Trend Interval
Use the Trend Interval tab in the PREFERENCES view to set the length of time used to plot the sparklines in the SYSTEM HEALTH DETAILS view. This interval applies to all metrics selected for this view.
The Example pane provides a sample of how the SYSTEM HEALTH view might look after the change.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Trend Interval tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
From the Interval list, select an interval.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
235
Chapter 24 System Health
236 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 25
Today's Statistics
Today's Statistics
The Todays Statistics portlet allows you to monitor current resource-use statistics and compare them to statistics collected during the same period one or more weeks in the past.
Additionally, you can monitor and compare the performance of multiple Teradata Database systems by adding a Todays Statistics portlet to the portal page for each monitored system.
The Teradata Viewpoint Administrator typically uses the Todays Statistics portlet to determine if users are getting value from their Teradata Database system. Data is presented in a table, allowing you to compare metric values collected recently (today and as far in the past as yesterday) to values collected during the same time period up to 999 weeks in the past. This view is refreshed once every 2 minutes with displayed values updated for each userselected metric.
About the Today's Statistics View
The TODAYS STATISTICS view displays a statistical analysis of system-resource use over a user-defined time period. A table lists the monitored metrics and compares current statistics to statistics from the same time period 1 or more weeks in the past. Metrics are grouped into categories so that metric data collected at the same frequency is displayed together in the view. The view is refreshed every 2 minutes.
Use the PREFERENCES view to select the system and metrics category to monitor, set metrics thresholds, and set the sampling interval. The heading for the second column depends on the selections made in the PREFERENCES view and how frequently data for each metric is collected. For example, the heading LAST 2 HOURS appears when the sampling interval is 2 hours and data for the previous 2 hours is available. The same heading could appear as SINCE AUG 20, 2:10 AM if the sampling interval is set to 2 hours and the data was last collected more than 2 hours ago.
Table columns provide the following information:
STATISTIC
The name of the monitored metric. This column can be sorted.
SINCE DATE, X AM/PM
The value of the metric at the date and time the data was last collected.
This column heading is used for metrics in the query category.
Teradata Viewpoint User Guide, Release 13.11
237
Chapter 25 Today's Statistics
LAST X HOUR(S)
The value of the metric from a user-specified number of hours ago, up to 999 hours in the past, until the current time.
SAME PERIOD X WEEK(S) AGO
The value of the metric over the same time period as LAST X HOUR(S), up to 999 weeks in the past.
SINCE X AM/PM TODAY/YESTERDAY
The value of the metric calculated from a user-specified hour of the day, either today or as far in the past as yesterday, until the current time.
SAME PERIOD X WEEK(S) AGO
The value of the metric over the same time period as SINCE X AM/PM TODAY/
YESTERDAY, up to 999 weeks in the past.
The Teradata Viewpoint Administrator must enable the appropriate collectors to display statistics in the TODAYS STATISTICS view. The Teradata Viewpoint Administrator also specifies the data-collection rate for each collector.
Today's Statistics View Metrics
Information about available metrics is provided in the tables. Use the Metrics tab in the
PREFERENCES view to select metrics for display in the TODAYS STATISTICS view.
The Teradata Viewpoint Administrator must enable the appropriate collectors to calculate and display statistical data in the TODAYS STATISTICS view. When the Canary Queries
Collector is the source for a metric, the data displayed is always the heartbeat query for the selected system.
System Metrics
The System Statistics collector gathers system metrics. System metrics are calculated as an average.
Metric
Active Sessions
AMP CPU Skew
AMP I/O Skew
AWT
CPU
System CPU
Total Disk Space
User CPU
Wait I/O CPU
Description
Number of sessions with active queries
Type
Number
Comparison of CPU use on the busiest AMP to the average AMP Percent
Comparison of disk use on the busiest AMP to the average AMP Percent
Number Average number of AMP worker tasks (AWTs) in use on each
AMP
Average node CPU use. CPU is calculated as the sum of the user CPU, system CPU, and wait I/O CPU usage percentages.
Percent
Average CPU time spent in system mode
Total disk space currently in use
Average CPU time spent in user mode
Average CPU time spent waiting for I/O
Percent
Number
Percent
Percent
238 Teradata Viewpoint User Guide, Release 13.11
Chapter 25 Today's Statistics
Query Metrics
The Query Count collector gathers query metrics.
Metric
Concurrency
Description
Average number of queries executing simultaneously during the hour
Type
Number
Average time (milliseconds) it takes to receive the first response Number First Response Time
Avg
First Response Time
Max
First Response Time
Min
First Response Time
Stddev
Query Count
Spool Usage Avg
Spool Usage Max
Spool Usage Min
Spool Usage Stddev
Maximum time (milliseconds) it takes to receive the first response
Minimum time (milliseconds) it takes to receive the first response
Standard deviation time (milliseconds) it takes to receive the first response
Number
Number
Number
Queries logged in dbc.QryLog, or the sum of Querycount from dbc.QryLogsummary during the sample period
Average amount of spool (bytes) the query used
Number
Number
Maximum amount of spool (bytes) the query used
Minimum amount of spool (bytes) the query used
Number
Number
Standard deviation of amount of spool (bytes) the query used Number
Performance Metrics
The Canary Queries collector gathers performance metrics. Performance metrics are calculated as an average.
Metric
Response Time
Rows Returned
Description
Average response time (milliseconds) of the selected canary query
Type
Number
Average number of rows returned by the selected canary query Number
About the Preferences View
The PREFERENCES view allows you to customize the portlet to monitor key metric trends to determine if you are getting value from your Teradata Database system. Only one system can be selected for each portlet instance.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
System
Select a Teradata Database system to monitor.
Teradata Viewpoint User Guide, Release 13.11
239
Chapter 25 Today's Statistics
Metrics
Select appropriate metrics, by category, for display in the view, and set values for highlighting metrics beyond a specified threshold. Metric values that exceed the threshold are displayed in the exception color.
Sampling Interval
Specify the time periods from which statistics are displayed. Set the number of hours (up to 999) prior to the current time, and define a comparison period up to
999 weeks in the past.
Selecting a System to Monitor
Use the System tab in the PREFERENCES view to select a Teradata Database system to monitor. Only one system can be selected for each portlet instance.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the System tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a system from the list.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Metric Thresholds
Use the Metrics tab in the PREFERENCES view to select a metrics category to monitor and to set thresholds to trigger the highlighting of values that exceed the thresholds.
To display statistical data in the TODAYS STATISTICS view, the Teradata Viewpoint
Administrator must first enable the appropriate Teradata Viewpoint collectors, including
Query count, System statistics, and Canary queries. Factors external to Teradata Viewpoint can also affect the collection and display of statistical data:
• Query Logging. The Teradata System Administrator must enable query logging on the
Teradata Database system.
• Data Collection. Query data must have been collected during all time periods selected on the Sampling Interval tab. Teradata Viewpoint collectors and query logging must have been enabled during all selected time periods in order for TODAYS STATISTICS data to display correctly.
1
2
3
4
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a metrics category.
240 Teradata Viewpoint User Guide, Release 13.11
Chapter 25 Today's Statistics
5
6
7
8
When the Canary Queries Collector is the source for a metric, the data displayed is always the heartbeat query for the selected system.
In the Metric column, select check boxes for the metrics you want to monitor in the view.
[Optional] For each metric selected, enter a Threshold value.
Values exceeding the threshold are highlighted in the view using the exception color.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting the Sampling Interval
Use the Sampling Interval tab in the PREFERENCES view to set the length of time used to calculate the statistics shown in the TODAYS STATISTICS view. The Sampling Interval applies to all metrics selected for display.
1
2
3
4
5
6
7
8
From the portlet frame, click to access the PREFERENCES view.
Click the Sampling Interval tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter the number of hours (up to 999) in the Statistics will be sampled during the last
X hour(s) field.
The value you enter determines the sampling interval for columns 2 and 3 of the
TODAYS STATISTICS view.
Enter the since sampling interval as follows:
a
Select the time of day that begins the interval.
b
Select Today or Yesterday from the list.
This value pair defines a start time up to 48 hours in the past for calculating column 4 of the TODAYS STATISTICS view.
Enter a value for the comparison period, up to 999 weeks in the past.
The value you enter defines the number of weeks in the past for calculating column 5 of the TODAYS STATISTICS view.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
241
Chapter 25 Today's Statistics
242 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 26
TVS Monitor
TVS Monitor
The TVS Monitor portlet allows you to view statistics on data temperature and storage grade of cylinders allocated in Teradata Virtual Storage (TVS). Statistics reflect current relationships between data temperature and storage grade, and historical trends in the management of storage grades based on data temperature.
Temperature and grade metrics can help assess storage requirements for your data, as well as fine tune the balance between disk performance and data-access demands. Suppose that statistics indicate an increasing proportion of hot data allocated to slow-grade storage. If fast storage is currently full, the Teradata System Administrator might consider upgrading to faster storage. If fast storage is not full, the Teradata System Administrator should verify that data migration and metrics collection have been enabled to ensure the slow migration of hot data over time.
About Storage Grade
Storage grade refers to the access rate capabilities of disk storage. Access speed is graded as slow, medium, or fast. Performance criteria for storage grade include physical speed of disk read-write access, electronics, and buffer memory.
Slow-grade storage involves more I/O time than fast-grade storage. Cylinders on physical disk drives are often associated with slow- and medium-grade storage. Cylinders on solidstate drives can be associated with fast-grade storage.
Typically, Teradata Virtual Storage optimizes cylinder allocation by reserving fast-grade storage for frequently accessed data, and slow-grade storage for infrequently accessed data.
Medium-grade storage is allocated for all other data-access frequencies.
About Data Temperature
Data temperature refers to the frequency of access to data stored in space types managed by
Teradata Virtual Storage (TVS). Data is considered hot, warm, or cold.
Data temperature is not static, but evolves over time, based on data-access patterns tracked by TVS. Data temperature increases the more often it is accessed, and decreases as access becomes less frequent.
Teradata Viewpoint User Guide, Release 13.11
243
Chapter 26 TVS Monitor
About Space Types
A space type is a classification of how disk space is allocated in the Teradata Database. Each space type, such as Perm, Spool, or Temp, is conventionally associated with a particular kind of stored data. For example, the Perm space type is disk space allocated to storage of tables, indexes, and permanent journals in the database, and typically is the largest space type in the system.
Teradata Virtual Storage reports on how space types are dynamically managed, according to patterns of data access and the grade of the storage device.
Selecting a View
You can select different views of the relationships between data temperature, storage grade, and space type.
1
Click one of the following:
•
to see the Cylinder Distribution by Grade view.
•
to see the Historical Cylinder Distribution by Grade view.
•
to see the Cylinder Allocation by Temperature and Grade view.
Viewing Cylinder Distribution by Grade
The Cylinder Distribution by Grade view displays graphs of the current proportional allocation of cold, warm, and hot data in slow-, medium-, and fast-grade storage, as well as the total number of cylinders assigned to each storage grade. The dashed line in each graph marks peak allocation for the storage grade in number and percent of cylinders, aggregated over the past 30 days.
1
2
3
Click the Slow Storage graph to display metrics on total slow cylinders, data temperatures, and peak allocation for slow storage.
Click the Medium Storage graph to display metrics on total medium cylinders, data temperatures, and peak allocation for medium storage.
Click the Fast Storage graph to display metrics on total fast cylinders, data temperatures, and peak allocation for fast storage.
Viewing Historical Cylinder Distribution by Grade
The Historical Cylinder Distribution by Grade view displays data temperature trends over time in each storage grade. The horizontal axis marks days and the vertical axis marks number of cylinders, in millions. You can change how temperature allocation metrics and total number of cylinders are displayed for a specific point in time.
1
2
Mouse over the graph to display the time marker.
Move the time marker with the mouse to change the display of day and hour.
244 Teradata Viewpoint User Guide, Release 13.11
Chapter 26 TVS Monitor
3
4
[Optional] Select a grade from the list to show proportion of cylinders, allocated by temperature, for the grade.
[Optional] Select a time period from the list to show proportion of cylinders, allocated by temperature, over the previous week or month.
Viewing Cylinder Allocation by Temperature and Grade
The Cylinder Allocation by Temperature and Grade view displays the current relationships between data temperature, storage grade, and space type. Change the display of the matrix and the Space Types by Cylinder Percentage table to see allocation metrics for a specific combination of temperature, grade, or space.
1
2
3
Select a cylinder-allocation space type from the list above the matrix to display a space type in the matrix and space-types table.
To filter the data shown in the space-types table, do one of the following:
a b
Click a row label (HOT, WARM, COLD) in the matrix to filter the space-types table by data temperature allocation.
Click a column label (SLOW, MEDIUM, FAST) in the bottom row of the matrix to filter the space-types table by storage grade.
c
Click any cell in the matrix to filter the space-types table by temperature-grade combination.
Select a temperature, grade, or temperature-grade combination from the list in the spacetypes table to filter space types by temperature, grade, or temperature-grade combination.
Teradata Viewpoint User Guide, Release 13.11
245
Chapter 26 TVS Monitor
246 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 27
Viewpoint Monitoring
Viewpoint Monitoring
The Viewpoint Monitoring portlet allows you to analyze resource usage trends on one or more Teradata Viewpoint servers or other managed servers. Trends are graphed according to metrics such as CPU, memory, and throughput within a specified time range.
The VIEWPOINT MONITORING view displays a graph with time on the horizontal axis, metric values on the vertical axis, and a different color for each metric sparkline. Sparklines show the average metric values as a line and have enhanced interactive features such as performance envelopes and information balloons that show the actual, minimum, and maximum metric values. Time-range buttons allow you to monitor current usage within the last hour or view usage trends over the last day, week, month, 3 months, or 6 months.
The PREFERENCES view allows you to select and organize metrics to display, set thresholds, and adjust the vertical axis range for each metric.
About the Viewpoint Monitoring View
The VIEWPOINT MONITORING view displays a graph that shows Teradata Viewpoint server resource usage. One or more Teradata Viewpoint servers can be monitored in the time range using the same or different metrics. Up to 10 metrics can be selected with different thresholds and vertical axis ranges. Each sparkline is assigned a color for every metric plotted in the graph.
The following list describes the features in this view:
Toolbar
Contains the time-range buttons. Select a time-range button to change the graph.
Graph
Plots the time range on the horizontal axis and the metric values on the vertical axis.
The sparkline is a colored line that represents the average metric values. Click a sparkline to see the highlighted sparkline and performance envelope. The performance envelope appears in a lighter shade of the sparkline color and represents the upper and lower metric values.
Time Range
Shows the range of time selected with the time-range button on the Toolbar. For example, if one hour is selected, times from the previous hour appear across the bottom of the graph.
Teradata Viewpoint User Guide, Release 13.11
247
Chapter 27 Viewpoint Monitoring
Metrics
Lists the metrics that are defined in the PREFERENCES view. Each metric has a color and a checkbox. The metric name appears along with the name of the system being monitored if there is more than one system enabled. Select a metric name to see the sparkline highlighted and surrounded by the performance envelope in the graph.
248
About Metric Thresholds
Set thresholds to highlight metric values that are outside normal operating ranges. Set maximum vertical-axis ranges to limit the range of values displayed for selected metrics.
After a metric threshold is set in the PREFERENCES view, the threshold line is displayed across the graph with the threshold value to the right of the graph when you select a metric sparkline. As you mouse over the selected sparkline, information balloons display the data point value and the minimum and maximum performance envelope values. The sparkline and information balloons change to red when the threshold is exceeded during the time range. Threshold settings are optional and can be set any time after the metric has been configured.
Teradata Viewpoint User Guide, Release 13.11
Chapter 27 Viewpoint Monitoring
Setting the Time Range
You can set the time range that is used to plot the graph.
1
In the toolbar, click a time-range button.
The portlet refreshes, and the graph is redrawn.
Removing Metrics from the Graph
You can disable metric sparklines.
1
Clear the color-coded checkbox on the metric line.
The metric sparkline disappears from the graph, but not from the list below the graph.
About the Toolbar
The toolbar allows you to choose the time range to display in the graph. The graph displays oldest data on the left and the most recent data on the right. For each metric, data is collected every 15 seconds and averaged according to the time range chosen.
Button
1 hour
1 day
1 week
1 mo
3 mo
6 mo
Description
Displays the metric values for the last hour, plotted by minute.
Displays the metric values for the last 24 hours, plotted by 15-minute periods.
Displays the metric values for the last week, plotted by hour.
Displays the metric values for the last month, plotted by 4-hour periods.
Displays the metric values for the last 3 months, plotted by 12-hour periods.
Displays the metric values for the last 6 months, plotted by day.
Teradata Viewpoint User Guide, Release 13.11
249
Chapter 27 Viewpoint Monitoring
Clicking a different time-range button causes the portlet to immediately recalculate and update the graph based on the data points collected for the metrics.
Viewpoint Monitoring Metrics
The following metrics are available to analyze resource usage.
CPU Metrics
Metric
CPU IDLE
CPU INVOLUNTARY
WAIT
CPU I/O WAIT
CPU SYSTEM
CPU USER
CPU USAGE
Description
Percent of total CPU time spent idle.
Percent of total CPU time spent waiting involuntarily.
Percent of total CPU time spent waiting for I/O.
Percent of total CPU time used by system processes.
Percent of total CPU time used by user processes.
Percent of CPU time used by user and system processes.
Type
Percent
Percent
Percent
Percent
Percent
Percent
Memory Metrics
Metric Description
MEMORY BUFFERS Amount of memory currently used as a disk buffer cache.
MEMORY CACHE Amount of cache memory.
MEMORY FREE
MEMORY TOTAL
Amount of idle memory.
Total amount of physical memory in the system.
Amount of memory used for swap space.
MEMORY TOTAL
SWAP
MEMORY USAGE Total amount of memory used by the system, calculated as the total memory minus the amount of free memory.
MEMORY VIRTUAL Amount of virtual memory used.
PROCS SLEEPING Number of processes in uninterruptible sleep.
PROCS WAITING
SWAP IN
SWAP OUT
Number of processes waiting for run time.
Amount of memory swapped in from disk per second.
Amount of memory swapped to disk per second.
KB
KB
KB
Type
KB
KB
KB
KB
Number
Number
KB/sec
KB/sec
Throughput Metrics
Metric
AVG. JOBS LAST
MINUTE
AVG. JOBS LAST FIVE
MINUTES
Description
Average number of processes running or able to run during the last minute.
Average number of processes running or able to run during the last 5 minutes.
Type
Number
Number
250 Teradata Viewpoint User Guide, Release 13.11
Chapter 27 Viewpoint Monitoring
Metric
AVG. JOBS LAST
FIFTEEN MINUTE
Description
Average number of processes running or able to run during the last 15 minutes.
BLOCKS RECEIVED Number of blocks received from a block device per second.
BLOCKS SENT
CONTEXT
SWITCHES PER
SECOND
Number of blocks sent to a block device per second.
Number of context switches per second.
INTERRUPTS PER
SECOND
Number of interrupts per second.
Type
Number
Number
Number
Number
Number
Metric Colors
A fixed set of 10 colors is used for the sparkline, performance envelope, and information balloon. Each metric you select is listed below the graph, and its color is determined by the position selected in the PREFERENCES view.
5
6
3
4
7
8
9
10
Metric Position Color Example
1
2
Description
Green
Blue
Gold
Aqua
Pink
Gray
Lime
Brown
Teal
Purple
About the Preferences View
The PREFERENCES view allows you to customize the appearance of the graph.
In this view, you can:
• Add metrics to monitor and display.
• Delete metrics that you do not want to monitor or display.
• Select the ENABLE check box to plot the metric in the graph.
• Clear the ENABLE check box to remove the metric from the graph, but still list it below the graph.
• Reorder metrics to change their position and color.
• Select systems to monitor and display.
Teradata Viewpoint User Guide, Release 13.11
251
Chapter 27 Viewpoint Monitoring
• Set a threshold that helps you track the performance of each metric.
• Set a vertical axis range to limit the maximum value to display.
From the portlet frame, click to access the PREFERENCES view.
Adding Metrics
You can add metrics that are plotted in the graph.
1
2
3
4
5
6
7
8
9
10
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Click to add a metric.
Up to 10 metrics can be listed.
Select a metric from the METRIC list.
[Optional] Select a system from the SYSTEM list.
The SYSTEM list is available when more than one Teradata Viewpoint server is enabled for the Data Collection Service.
[Optional] Enter a THRESHOLD value.
You can change the THRESHOLD settings for the metric at any time.
[Optional] Enter a VERTICAL AXIS RANGE value.
You can change the VERTICAL AXIS RANGE settings for the metric at any time.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Changing the Metrics Display Order
You can change the order of the metrics that are listed below the graph. Reordering the metric list affects which color is assigned to the metric.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
On the metric row, click and drag the row to its new location.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
252 Teradata Viewpoint User Guide, Release 13.11
Chapter 27 Viewpoint Monitoring
Enabling and Disabling Metrics
You can enable and disable metrics from the graph. When a metric is disabled, the metric sparkline is not plotted in the graph; however, data points are still being collected, so the metric is included in the list under the graph and can be reactivated at a later time.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Select the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do one of the following:
• Clear ENABLE on the metric line to disable the metric.
• Select ENABLE on the metric line to enable the metric.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Deleting Metrics
You can delete metrics from the graph. When a metric is deleted in the PREFERENCES view, the metric sparkline is not plotted in the graph, and the metric is not included in the list under the graph.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Select the Metrics tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Click on the metric row of the metric you want to delete.
The metric row disappears.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
253
Chapter 27 Viewpoint Monitoring
254 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 28
Workload Designer
Workload Designer
The Workload Designer portlet allows you to define, control, balance, and refine rules for managing workloads. Use the following features to manage workloads:
• Rulesets which are collections of related filters, throttles, events, states, and workload rules
• States which cause actions when a specific combination of planned environment and health condition occur
• Sessions limits, including query sessions limits, utility sessions limits, and utility limits
• Filters which reject queries
• Throttles which limit queries
• Exceptions which cause actions when one or more specific events occur
• Classification criteria which determine which queries use which rules
The Workload Designer portlet is a full-width portlet that cannot be minimized.
The WORKLOAD DESIGNER view provides controls to select a system and create, edit, view, and perform other actions on rulesets on that system.
The Workload Designer Ruleset views provide controls that allow you to refine settings for a system, states, sessions, filters, throttles, workloads, and exceptions.
About the Workload Designer View
The WORKLOAD DESIGNER view shows summary information about rulesets. Items in the options list depend on whether you are the ruleset owner. If a ruleset is locked by someone else, you have fewer options than if you are the ruleset owner. The Working,
Ready, and Active sections contain the following options:
Working
Names and descriptions of rulesets that are being edited. In Working, you can create and import rulesets. Rulesets in Ready can be copied to Working for editing.
Rulesets in Working can also appear in Ready and Active.
Ready
Rulesets that have been saved to the production system, but are not active. A ruleset must be in Ready before it is copied to Active. The Active ruleset cannot be deleted from Ready.
Teradata Viewpoint User Guide, Release 13.11
255
Chapter 28 Workload Designer
Active
Active ruleset on the production system. The only option available in the options list, if you have permissions, is to deactivate the ruleset.
Ruleset background color indicates synchronization between the Working and Ready sections. Gray in both sections indicates that the same version of the ruleset is in both sections. Blue indicates that the ruleset has been modified in the Working section since it was last saved to the Ready section.
TASM ruleset migration is performed automatically during the Teradata Database upgrade.
About Ruleset Options
The WORKLOAD DESIGNER view shows individual rulesets and specific options for each ruleset based on the permissions assigned to your role and the section in which the ruleset is located.
Working
The Working section contains rulesets that are being edited. Any ruleset listed in Working can also be listed in Ready and Active. Available options are:
• Make Ready. Copies the ruleset to the production server and to the Ready section.
• Make Active. Makes the ruleset the active ruleset on the production server. Copies the ruleset to the Ready and Active sections.
• View. Opens the ruleset for viewing.
• Edit. Opens the ruleset for editing.
• Show All. Lists all ruleset attributes on one page.
• Lock. Locks the ruleset so only the lock creator can edit the ruleset.
• Unlock. Unlocks the ruleset so others can edit the ruleset.
• Clone. Creates a copy of the ruleset. This option is useful if you want to use an existing ruleset as a base or template to create a ruleset.
• Export. Exports the ruleset XML file so you can view the file in a browser or save the file to a location you specify. Use with the Import button to copy a ruleset from one system to another.
• Delete. Removes the ruleset from the Working section.
Ready
The Ready section lists rulesets saved to the production server. Someone else can create a ruleset and add it to the Ready section. Then, from the Ready section, you can select Copy to
Working Rulesets and edit the ruleset. Available options are:
• Activate. Makes the ruleset the active ruleset on the production server.
• Copy to Working Rulesets. Copies the ruleset to the Working section.
• Delete. Removes the ruleset from the production server.
Active
The Active section contains the active ruleset on the production server. If you have permission, the only available option is to Deactivate the current ruleset.
256 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Creating a Ruleset
A ruleset is a complete collection of related filters, throttles, events, states, and workload rules. You can create multiple rulesets, but only one ruleset is active on the production server at a time. After creating a ruleset, you can specify settings, such as states, sessions, and workloads, using the toolbar buttons. New rulesets are automatically locked so only the owner can edit the ruleset.
1
2
3
4
5
6
7
8
9
10
11
12
From the WORKLOAD DESIGNER view, select a system from the list.
Click Create.
Enter a ruleset name.
[Optional] Enter a description up to 80 characters long.
Click Save.
[Optional] Specify settings using the tabs in the General view, such as Intervals and
Bypass.
[Optional] Click States and create a state matrix.
[Optional] Click Sessions and create any of the following:
•
Query Sessions
•
Utility Limits
•
Utility Sessions
[Optional] Click Filters and create a filter.
[Optional] Click Throttles and create a system throttle.
[Optional] Click Workloads and create a workload.
[Optional] Click Exceptions and create an exception.
Editing a Ruleset
In the WORKLOAD DESIGNER view, you can edit ruleset properties. The Teradata
Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action. Without permission to edit the ruleset, the menu option is View and the ruleset view is read-only.
1
2
3
4
From the WORKLOAD DESIGNER view, select a system from the list.
In the local Working section, click the ruleset name.
Specify settings using the toolbar buttons and tabbed views.
Click Save after making changes in each view.
Cloning a Ruleset
Cloning a ruleset makes an exact copy of the ruleset, except for the name and description.
Cloning is a convenient way to create a ruleset using the criteria of an existing ruleset as a base.
1
From the WORKLOAD DESIGNER view, select a system from the list.
Teradata Viewpoint User Guide, Release 13.11
257
Chapter 28 Workload Designer
2
3
4
5
6
7
In the local Working section, click the ruleset and click Clone.
An exact copy of the ruleset is made, and the General view appears.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Click Save.
Specify additional settings using the toolbar buttons and tabbed views.
Click Save.
Deleting a Ruleset
Deleting a ruleset removes the ruleset and all associated information. The Teradata
Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action.
1
2
3
From the WORKLOAD DESIGNER view, select a system from the list.
In the local Working section or the system Ready section, click the ruleset , and click
Delete.
Click Delete.
The ruleset is deleted from the section.
Importing a Ruleset
The import and export options can be used to copy a ruleset from one system to another. The
Teradata Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action. Only rulesets exported from Workload Designer and a Teradata
Database of the same release can be imported.
1
2
3
4
5
6
From the WORKLOAD DESIGNER view, select a system from the list.
Click Import.
Enter a name for the imported ruleset.
Click Browse.
Locate and select the saved ruleset file.
Note: Exported ruleset files might be stored in the download area configured for your browser.
Click Save.
Exporting a Ruleset
The import and export options can be used to copy a ruleset from one system to another. The
Teradata Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action. Only rulesets exported from Workload Designer and a Teradata
Database of the same release can be imported.
1
From the WORKLOAD DESIGNER view, select a system from the list.
258 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
2
3
In the local Working section, click the ruleset , and click Export.
Click Save.
The ruleset file is saved to your download area or the location you specify, depending on your browser settings.
Showing All Criteria in a Ruleset
You can display a read-only summary of all settings and state-specific values for a single ruleset.
1
2
3
From the WORKLOAD DESIGNER view, select a system from the list.
In the local Working section, click the ruleset , and click Show All.
[Optional] Do any of the following:
• Click Collapse All to hide all sections of the view.
• Click Print to print the full summary.
• Click the section name to show or hide individual sections of the view.
Copying a Ruleset to the Production System
When you are finished editing a ruleset in the Working section, copy the ruleset to the
Ready section on the production system. The Teradata Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action. After copying a ruleset to the Ready section, you can make the ruleset the active ruleset on the production system.
1
2
From the WORKLOAD DESIGNER view, select the production system containing the ruleset you want to copy.
In the Working section, click the ruleset , and click Make Ready.
Activating a Ruleset
Activating a ruleset copies the ruleset to the active state on the selected system. Only one ruleset is active on the system at a time. A ruleset activated from the local Working section is copied to the Ready section before being made active. The Teradata Viewpoint
Administrator must grant your role permission to edit rulesets so you can complete this action.
1
2
From the WORKLOAD DESIGNER view, select a system from the list.
From a ruleset in the Working or Ready section, click the ruleset , and click Make
Active.
Deactivating a Ruleset
Deactivating a ruleset removes the ruleset from the active state on the selected system. The
Teradata Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action.
1
2
From the WORKLOAD DESIGNER view, select a system from the list.
In the system Active section, click the ruleset , and click Deactivate.
About Ruleset Locks
An exclusive lock can be placed on a ruleset so that the ruleset cannot be edited, deleted, or otherwise modified except by the owner of the lock. A ruleset is automatically locked by the
Teradata Viewpoint User Guide, Release 13.11
259
Chapter 28 Workload Designer user when it is created and each time changes to the ruleset are saved. Use the WORKLOAD
DESIGNER view to lock and unlock rulesets. The Teradata Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action.
The WORKLOAD DESIGNER view displays ruleset lock status:
Option Description
Locked.
Unlocked.
The ruleset views display ruleset lock status:
Option Description
Unlocked.
Locked by the current user.
Locked by another user.
Locking or Unlocking a Ruleset
Locking a ruleset prevents others from editing or deleting the ruleset. The Teradata
Viewpoint Administrator must grant your role permission to edit rulesets so you can complete this action.
1
2
From the WORKLOAD DESIGNER view, select a system from the list.
In the local Working section, click the ruleset , and click Lock or Unlock.
About the Ruleset General View
The ruleset General view displays general attributes of a ruleset and appears after you click the General button on the ruleset toolbar. The General view appears by default when creating, editing, cloning, or viewing a ruleset. This view contains the following tabs:
General
Ruleset name and description.
Intervals
Collection and reporting intervals.
Option Description Default
Value
Minimum
Value
60 seconds 1 second
Maximum
Value
3600 seconds Event Interval How often event thresholds are checked.
Dashboard
Interval
How often workload statistics are collected.
60 seconds 1 second 600 seconds
260 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Option Description Default
Value
Minimum
Value
600 seconds 1 second
Maximum
Value
3600 seconds Logging Interval How often workload and exception logs are written.
Exception
Interval
How often exception thresholds are checked.
60 seconds 1 second 3600 seconds
Blocker
Settings for responding to throttled blockers. The log is located at DBC.TDWMEventLog.
Selecting Log only logs the blocker. If Abort or Release is selected, the action is logged after the abort or release occurs.
Option
Block
Cycles
Description
The number of intervals over which the query must be blocked before the specified Block Action is taken.
Default
Value
Off
Values
2
3
Off
1
Block
Action
The action to perform in response to query blocking.
Log
Log
Abort (abort query)
Release (release query from queue)
Activation
Features that are available when the ruleset is activated. Events and States is a required field and can be the only field selected.
Bypass
Users, accounts, and profiles whose queries are allowed to bypass system filters and throttles.
Option
Source
Type
Description
Type of query source (you can select one or more items from each available source type):
Usernames (usernames dbc and tdwm cannot be removed)
Account Names
Account Strings
Profiles
Filter
Items
Bypass
Limits your choices in the source list.
Lists the names of all available query sources of the selected type.
Selected sources, listed by source type. Queries from each of the listed query sources will bypass system filters and throttles for this ruleset.
Teradata Viewpoint User Guide, Release 13.11
261
Chapter 28 Workload Designer
Defining System-Level Bypass Settings
A system-level bypass is a collection of users, accounts, and profiles that are not filtered or throttled at the system level. The dbc and tdwm usernames are listed automatically and cannot be changed, renamed, or removed.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click General.
Click the Bypass tab.
Select a Source Type from the list to exclude it from system-level filters and throttles.
[Optional] Enter a filter string in the Filter box to limit your choices.
Select a source from the list or select multiple sources by using Ctrl or Shift.
Click Add Selection to add your selections to the bypass list.
[Optional] Repeat steps 4 through 7 to add sources to the bypass list.
[Optional] Mouse over a source in the bypass list, and click to include it in systemlevel filters and throttles.
Click Save.
About Ruleset States
A state is the intersection of a health condition and a planned environment. A health
condition is composed of unplanned events and a planned environment is composed of planned events. Creating states provides greater control over how the system allocates resources. When a health condition and a planned environment intersect, the resulting state triggers system changes.
Use the state matrix to create and organize states for a ruleset. With the state matrix, create states so one ruleset can respond to a range of different system conditions. For example, if you have many system users on weekdays, but run batch jobs on the weekend, allocate system resources differently during the week than you do on weekends by creating two planned environments: Weekdays and Weekends.
In the state matrix, create planned events, unplanned events, health conditions, planned environments, and corresponding states specific to your business situation. Update the state matrix at any time to reflect business, system, or priority changes.
The Normal health condition, Always planned environment, and Base state are defaults. The defaults apply unless planned or unplanned events occur, triggering other configured states.
The defaults cannot be deleted or moved within the state matrix. Like any state, the Base state can be used in multiple cells of the matrix.
262 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Any states you create use the default settings. The default settings can be viewed and edited on the state-specific settings tabs in workloads, filters, throttles, query sessions, and utility limits. You can override the default settings by entering new values on the state-specific settings tabs.
Using only a few states in the state matrix reduces maintenance time. However, consider adding states to the matrix to manage the following situations:
• Consistent, peak workload hours or days where priority management must be strictly assigned and enforced.
• Load or query times where priority tasks must finish within a specific time frame.
• Conditions where resources must be managed in a different way, such as giving higher priority to critical work when system health is degraded.
Creating a State Matrix
If you have created a ruleset, you can create a state matrix for the ruleset. The state matrix instructs the system as to which predefined state to use when a specific combination of planned environment and health condition exists.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click States.
The state matrix appears. For a new ruleset, the default state matrix consists of one cell.
Define at least one health condition.
Define at least one unplanned event.
Unplanned events are system events, user-defined events, or a combination of these events.
Drag the unplanned event to the appropriate health condition so the event triggers the health condition.
Define at least one planned environment.
Define at least one planned event.
Planned events are period events, user-defined events, or a combination of these events.
Drag the planned event to the appropriate planned environment so the event triggers the environment.
Define at least one state.
For each cell in the state matrix, drag and drop a state from the States list into the appropriate cell of the matrix. When the defined combination occurs, the state is triggered.
Teradata Viewpoint User Guide, Release 13.11
263
Chapter 28 Workload Designer
264
11
Click Save.
Defining Health Conditions
Health conditions define levels of system health and are used to reallocate system resources when an event degrades the system. When at least one unplanned event occurs, a health condition can be triggered. The default health condition is Normal, and it is used if no other health conditions are triggered. The Normal health condition always remains at the top. The lowest severity is listed at the top. The highest priority is at the bottom. If multiple unplanned events are active at the same time, the health condition with the highest severity is triggered.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Health Conditions, and click .
A health condition is added with the default name newCond.
Mouse over the health condition, and click .
Enter a name for the health condition.
Enter a minimum duration, in minutes, for the health condition.
The event that triggers the change to the health condition must remain above the trigger threshold for the minimum duration you enter before the health condition goes into effect. Setting minimum duration prevents short incidents of an event from triggering a change in the health condition.
Click OK.
[Optional] If there are more than two health conditions, click the health condition name, and drag the name to reorder.
[Optional] To delete a health condition, mouse over a health condition name, and click
.
Click Save.
Defining System Events in the State Matrix
System events are unplanned events, such as a down node. Incorporating system events into health conditions within the state matrix gives you greater control over what actions
Teradata Database takes when unexpected events occur. To create an event that only sends out a notification, create the event, but do not assign it to any unplanned environment.
When the event occurs, the notification action you specified is triggered.
Note: Some event types give you the option to set qualification time. Qualification time is the length of time the specified event must last to be recognized as an event. Setting qualification time prevents very short incidents of an event from being acknowledged as events.
1
2
3
4
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Unplanned Events, and click .
Click next to Available Events.
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
5
6
7
8
9
10
11
12
13
Select System Event.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select an Event Type from the list:
• Node Down. Specify the maximum allowed number of nodes down in a clique as a percentage between 0 and 100. The default is 24%.
• AMP Fatal. Specify the maximum number of AMPs reported as fatal at system startup.
• Available AWT's. Specify the minimum number of AWTs available on an AMP. Set a qualification time in minutes and seconds. (Release 13.10 and later.)
• AWT Limit. Specify the maximum number of AWTs in use on an AMP. Set the number of AMPs that must exceed the limit. Set a qualification time in minutes and seconds. (Release 12.0 and Release 13.0.)
• Gateway Fatal. Specify the maximum number of gateways reported as fatal at system startup.
• PE Fatal. Specify the maximum number of PEs reported as fatal at system startup.
• Flow Control. Specify the maximum allowed number of AMPs in flow control. Set a qualification time in minutes and seconds.
Maximum and minimum limits must be positive integers.
[Optional] Under Notifications, enable any of the following actions for the start or end of the event:
• Send Alert. Select the Teradata Alerts to run.
• Run Program. Select the program registered with Teradata Alerts to run.
•
Post to QTable. Enter a string to post to the QTable at the start or end of the event.
This option is not integrated with Teradata Alerts.
Click OK.
[Optional] Drag the event under a specific health condition.
Click Close.
Click Save.
Defining Planned Environments
Planned environments reallocate system resources during scheduled times. Planned environments are triggered when at least one planned event occurs. The default planned environment is Always and it cannot be deleted or moved. The order of precedence is from lowest to highest, reading from left to right. The planned environment with the highest precedence is activated if multiple planned events are active at the same time.
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Planned Environments, and click .
A planned environment is added with the default name newEnv.
Teradata Viewpoint User Guide, Release 13.11
265
Chapter 28 Workload Designer
4
5
6
7
8
9
Mouse over the planned environment, and click .
Enter a name.
Click outside the name.
If there are more than two planned environments, click a planned environment name, and drag the name to the left or right in the list to change the order of precedence.
[Optional] To delete a planned environment, mouse over the environment name, and click .
You cannot delete the Always environment.
Click Save.
Defining Period Events in the State Matrix
Period events are planned events occurring on specific days and times, such as month-end financial processing. To create an event that only sends out a notification, create the event, but do not assign it to any planned environment. When the event occurs, the notification action you specified is triggered.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Planned Events, and click .
Click next to Available Events.
Select Period Event.
Enter a name.
[Optional] Enter a description up to 80 characters long.
[Optional] Select Create New Corresponding Planned Environment to automatically create a planned environment that this event triggers.
Select Day of Week or Day of Month, and click a single day or multiple days.
[Optional] Select Month of Year, and click one or more months.
[Optional] Select Start Time, and enter a start and end time.
[Optional] Select Wrap around midnight to have a time range spanning midnight for a period event.
[Optional] In Notifications, select any of the following:
• Send Alert. Specify the Teradata Alerts to trigger at the start or end of the event.
•
Run Program. Specify the Teradata Alerts registered programs to trigger at the start or end of the event.
•
Post to QTable. Enter a string to post to the QTable at the start or end of the event.
(This option is not integrated with Teradata Alerts.)
Click OK.
[Optional] Drag the event under a planned environment.
266 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
16
17
Click Close.
Click Save.
Wrap Around Midnight Example
When creating period events, use the Wrap around midnight option to have a time range spanning midnight.
If the From time of a period event is later than the To time, two time segments are available: midnight until the To time and the From time until midnight. When Wrap around
midnight is not selected, the period event is in effect for segment 1 and segment 2 on each specified day. When Wrap around midnight is selected, the event is in effect for segment 2 on each specified day and for segment 1 on each day following the specified day.
For example, specify that a period event occurs on Mondays and Tuesdays with a From time of 17:00 and a To time of 08:00.
If Wrap around midnight is not selected: midnight—08:00 (time segment 1)
Monday
Yes
08:00—17:00
17:00—23:59 (time segment 2)
No
Yes
If Wrap around midnight is selected: midnight—08:00 (time segment 1)
08:00—17:00
17:00—23:59 (time segment 2)
Monday
No
No
Yes
Tuesday
Yes
No
Yes
Tuesday
Yes
No
Yes
Wednesday
No
No
No
Wednesday
Yes
No
No
Defining User-Defined Events in the State Matrix
User-defined events can be planned or unplanned. To create an event that only sends out a notification, create the event, but do not assign it to any planned or unplanned environment. When the event occurs, the notification action you specified is triggered.
1
2
3
4
5
6
7
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Unplanned Events or Planned Events, and click .
Next to Available Events, click .
Select User Defined Event.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Teradata Viewpoint User Guide, Release 13.11
267
Chapter 28 Workload Designer
268
8
9
10
11
12
13
14
15
[Optional] If you are creating the user-defined event as a planned event, you can select
Create New Corresponding Planned Environment to automatically create a planned environment that this event triggers.
Under Activate/Deactivate Event, copy the appropriate SQL request text.
Paste the text into an SQL script.
[Optional] Under Notifications, do any of the following:
• Send Alert. Specify the Teradata Alerts to trigger at the start or end of the event.
•
Run Program. Specify the Teradata Alerts registered programs to trigger at the start or end of the event.
•
Post to QTable. Enter a string to post to the QTable at the start or end of the event.
(In Release 12.0, you can select the Post to QTable option, but the text box is not available.)
Click OK.
[Optional] Do one of the following:
• For unplanned events, drag the event under a health condition.
• For planned events, drag the event under a planned environment.
Click Close.
Click Save.
Defining Event Combinations in the State Matrix
An event combination is a mix of two or more different events, such as period, system, and user-defined events. Event combinations can be planned or unplanned. To create an event that only sends out a notification, create the event, but do not assign it to any planned environment. When the event occurs, the notification action you specified is triggered.
1
2
3
4
5
6
7
8
9
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over Unplanned Events or Planned Events, and click .
Next to Available Events, click .
Select Event Combination.
Enter a name.
[Optional] Enter a description up to 80 characters long.
[Optional] If you are creating the combination event as a planned event, you can select
Create New Corresponding Planned Environment to automatically create a planned environment that this event triggers.
Add available events and operators to build the formula representing the combination of events that must occur to trigger the event. For example, Node Down OR Batch
Processing, or NOT PEI.
When creating event combinations, avoid placing two operators or two events next to each other. When an event combination is valid, the background of the Event
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
10
11
12
13
14
Combination Formula text box is white. If a combination is invalid, the background is orange.
[Optional] In Notifications, select any of the following:
• Send Alert. Specify the Teradata Alerts to trigger at the start or end of the event.
• Run Program. Specify the Teradata Alerts registered programs to trigger at the start or end of the event.
• Post to QTable. Enter a string to post to the QTable at the start or end of the event.
(This option is not integrated with Teradata Alerts.)
Click OK.
[Optional] Do one of the following:
• For unplanned events, drag the event under a health condition.
• For planned events, drag the event under a planned environment.
Click Close.
Click Save.
Defining States in the State Matrix
You can create a state to control how resources are allocated in different health condition and planned environment combinations.
1
2
3
4
5
6
7
8
Edit or create a ruleset.
From the ruleset toolbar, click States.
Mouse over States at the top of the states list and click .
A state is added to the list with the default name newState.
Mouse over the state name, and click .
Enter a name.
Click outside the name.
[Optional] To edit a state, mouse over the state name, and click .
Click Save.
Mapping States in the State Matrix
After creating a state, you can map it in the state matrix to control how resources are allocated when different health condition and planned environment combinations occur. By default, the cell in the upper left corner of the state matrix is assigned the Normal health condition, the Always environment, and the Base state. This cell cannot be changed. All other cells in the matrix must be associated with a single state. Any state, including the Base state, can be used in multiple cells.
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click States.
To map a state in the state matrix, drag and drop a state name from the States list to a matrix cell.
Teradata Viewpoint User Guide, Release 13.11
269
Chapter 28 Workload Designer
270
4
5
[Optional] To view details about a state, mouse over the state name in the States, list and click .
Click Save.
Mapping Events in the State Matrix
If you have created events in the state matrix, you can combine the events with health conditions and planned environments for greater control of Teradata Database. Create the health conditions and planned environments that you need before mapping events in the state matrix.
1
Edit or create a ruleset.
2
3
From the ruleset toolbar, click States.
Map a planned event:
4 a
Mouse over Planned Events, and click .
b
In Available Events, drag and drop an event from the list into one or more of the
Planned Environment columns.
c
[Optional] To create a planned event, click next to Available Events.
To map an unplanned event:
5
6 a b
Mouse over Unplanned Events, and click .
In Available Events, drag an event from the list and drop it into one or more of the
Health Condition rows.
c
[Optional] To create an unplanned event, click next to Available Events.
Click Close.
Click Save.
Deleting States from the State List
If a state is not used in the state matrix, you can remove it from the State list.
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
1
2
3
4
Edit or create a ruleset.
From the ruleset toolbar, click States.
In the State list, mouse over a state name, and click .
Click Save.
About Classification Settings
Workload Designer provides a common classification process for workloads, filters, throttles, query sessions, and utility sessions. Classification determines which queries use which rules. Teradata Database detects classification criteria before executing queries. The goal in creating a useful classification scheme is to meet business goals and fine-tune control of Teradata Database.
Modifications to the classification settings can be made in response to data monitoring, regular historical analysis, or changes. For example, classification groups may need to be created, or existing groups modified, if an application is added, two production systems are consolidated, or service-level goals are missed.
About Classification Criteria
The following classifications are available:
• Request source (Where does the query come from?)
• Target (What is the query acting on?)
• Query characteristics (What is the query composed of?)
• Query band (What metadata is attached to the query?)
• Utility (Which utility submitted the query?)
A good approach to using classification is to first use request source to determine where the query is coming from. Often, the account string is selected, but other options include username, account name, or client IP address. If you need a more detailed level of classification, establish where the query data is located, such as a database, table, or view. To narrow classification further, select query characteristics, query bands, or utilities. For utilities, use the check boxes to select the specific utilities you want to include.
For example, you could create a filter and add the request source classification to reject all queries from the Finance department when the Red state is in effect. To further refine the filter, add the query characteristic classification to filter out all requests from the finance department that are estimated to run longer than 10 seconds. (You include classification items in a filter to reject those items.) The query characteristic setting is added to the request source setting already in place.
All classification settings for a workload, filter, throttle, query session, utility session, or utility limit are listed on the Classification tab.
Teradata Viewpoint User Guide, Release 13.11
271
Chapter 28 Workload Designer
272
On the Classification tab:
• Request source and target criteria can have a single criteria or several. In Release 13.10, if a criteria group has more than one criteria, then within the group the criteria are joined by AND. In Release 12.0 and Release 13.0, the criteria are joined by OR.
• If request source and target groups exist together, they are joined by an AND in Release
13.10. In Release 12.0 and Release 13.0, a button can be used to join the groups by AND or by OR.
• Within a query characteristics criteria group, if the Statement Type parameter is enabled, it is joined to any additional parameters with an AND. In Release 13.10, if there is more than one parameter, they are joined by an AND. In Release 12.0 and Release 13.0, a button can be used to join the parameters by AND or by OR.
About Request Source Classification Type
The request source classification type establishes which username, account name, account string, profile, application, client IP address, or client ID is making the request.
Consider the following when using request source to classify information:
• A source type can only be used once per rule. After a source type is used, it no longer appears in the menu.
• A match string must be an exact match. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list. You can also select items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Adding Request Source Classification Type
You can classify filters, throttles, workloads, utility sessions, and query sessions by request sources such as account name or client IP address.
In Release 12.0 and Release 13.0, when adding the request source classification type to filters and throttles, an Add button is available instead of the Include and Exclude buttons. In
Release 12.0, when adding the request source classification type to workloads, all specified criteria can be either included or excluded. In Release 12.0, wild cards are not available in
Match String. In Release 12.0 and Release 13.0, the request source types available for filters and throttles are Username, Profile, Account Name, and Account String.
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
1
2
3
4
5
6
Click Filters, Throttles, Workloads or Sessions (Query Sessions or Utility Sessions tab) and select the name of a rule or create a rule.
Click the Classification tab.
Do one of the following:
• From the Add Classification Criteria list, click Request Source, and click Add.
• Click Add Criteria if you are classifying a query session.
• Select an existing request source criteria.
Select a Source Type from the list.
Do at least one of the following:
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. (Wildcards in match strings are supported in Release 13.0 and later.) Use the Match String Include and Exclude buttons to add the match string to a list.
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria. (In Release 12.0 and Release 13.0, use the Add button.)
Click OK.
About Target Classification Type
The target classification type establishes the query data location.
Consider the following when using target to classify information:
• Available target types include database, table, macro, view, or stored procedure. If table, macro, view, or stored procedure is selected, a database selection list appears. A target type can only be used once per rule. After a target type is used, it no longer appears in the menu.
• Optionally, each selected target item can have subcriteria. For example, if you select a database as the target, you could add subcriteria so that it only applies if you are performing a full table scan. If you select two or more subcriteria, they must all be present for the classification setting to be used. The icon appears next to target items containing subcriteria.
• A match string must be an exact match. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list. You can also select items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Adding Target Classification Type
The target classification type specifies the query data location. You can classify filters, throttles, or workloads by targets such as database, table, or stored procedure. You can add subcriteria in Release 13.10. If you add multiple subcriteria to a single item, all subcriteria conditions must be true in order for the query to be classified into the rule. In Release 12.0
and Release 13.0, when adding the target classification type to filters and throttles, an Add button is available instead of the Include and Exclude buttons. In Release 12.0, when adding the target classification type to workloads, all specified criteria can be either included or excluded. In Release 12.0, wild cards are not available in Match String.
Teradata Viewpoint User Guide, Release 13.11
273
Chapter 28 Workload Designer
1
2
3
4
5
6
7
8
9
Click Filters, Throttles, or Workloads and select the name of an existing item or create one.
Click the Classification tab.
Do one of the following:
• From the Add Classification Criteria list, click Target, and click Add.
• Select an existing target criteria.
Select a Target Type from the list.
[Optional] Select a database from the list for Table, View, Macro, or Stored Procedure target types.
Do at least one of the following:
• Enter a Match String and use the Match String Include and Exclude buttons to add the match string. A match string can contain ? to match exactly one character or an * to match zero or more characters.
• Select Items from the list and use the Items Include and Exclude buttons to add the items.
[Optional] To add subcriteria, mouse over an item in the Selected list, click , and choose from the following:
• Full Table Scan. Include or exclude full table (all row) scans.
•
Join Type. Select a type, such as No Join or Any Join.
• Minimum Step Row Count. Set minimum rows at each step.
• Maximum Step Row Count. Set maximum rows at each step.
•
Minimum Step Time. Set minimum time at each step.
If you specified subcriteria, click Apply.
The icon appears next to target items containing subcriteria.
Click OK.
About Query Characteristic Classification Type
The query characteristic classification type describes a query by answering such questions as what does the query do and how long will the query run.
Consider the following when using query characteristics to classify information:
• After a characteristic is selected, its value can be edited.
• Many characteristics have minimum and maximum values that can be set independently.
You can set all values above the minimum, below the maximum, or between a minimum and a maximum.
• Query characteristic classification and utility classification are mutually exclusive. If you use one, the other option is not available.
• You can have one query characteristic classification per rule.
• If you select Join Type, you can choose from No Join, Any Join, Product Join, No
Product Join, Unconstrained Product Join, and No Unconstrained Product Join.
274 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Adding Query Characteristic Classification Type
You can classify filters, throttles, or workloads by query characteristics.
1
2
3
4
5
Click Filters, Throttles, or Workloads and select the name of an existing item or create one.
Click the Classification tab.
Do one of the following:
• From the Add Classification Criteria list, click Query Characteristics, and click
Add.
• Select an existing query characteristics criteria.
Choose from the following query characteristics criteria:
• Statement Type. Click DDL, DML, or SELECT.
• AMP Limits. Include or exclude queries that use all AMPs. (Available for filters in
Release 13.10.)
•
Step Row Count. Set minimum and maximum rows at each step.
• Final Row Count. Set minimum and maximum rows in the result set.
• Estimated Processing Time. Set minimum and maximum estimated processing time. (Longer or more complex queries have less accurate estimates.)
• Minimum Step Time. Set a minimum time at each step.
•
Join Type. Use the Include Only list to select one of the following join types: No
Join, Any Join, Product Join, No Product Join, Unconstrained Product Join, or No
Unconstrained Product Join.
• Full Table Scan. Choose to include or exclude full table (all row) scans.
Click OK.
About Query Band Classification Type
The query band classification type describes the query band data attached to a query.
Consider the following when using query band to classify information:
• A name must be selected from the Name list or entered into the box.
• After picking a name, one or more values must be specified. The value is selected from the Previously Used Values list or entered into the New Value box. Multiple values can be selected for the same name.
• After a name and value are specified, the Include and Exclude buttons are available.
• Multiple included query band key and value pairs are connected with "AND."
• Multiple excluded query band key and value pairs are connected with "OR."
Adding Query Band Classification Type
A query band contains name and value pairs that use predefined names (on Teradata
Database) or custom names to specify metadata, such as user location or application version. The query band classification type describes the query band data attached to a query.
In Release 12.0 and Release 13.0, when adding the query band classification type to filters and throttles, an Add button is available instead of the Include and Exclude buttons. In
Teradata Viewpoint User Guide, Release 13.11
275
Chapter 28 Workload Designer
Release 12.0, when adding the query band classification type to workloads, all specified criteria can be either included or excluded.
1
2
3
4
5
6
7
Click Filters, Throttles, or Workloads and select the name of an existing item or create one.
Click the Classification tab.
Do one of the following:
• From the Add Classification Criteria list, click Query Band, and click Add.
• Select an existing query band criterion.
Do one of the following:
• Select a predefined query band name from the list.
• Enter a name.
Select a Previously Used Value or enter a New Value.
You must select a name and a value.
Use the Include and Exclude buttons to create your classification criteria.
Click OK.
About Utility Classification Type
The utility classification type identifies which utility submitted the query.
Consider the following when using utility to classify information:
• Available utility types include FastLoad, FastExport, MultiLoad, and Archive/Restore.
Select a top level utility such as FastExport or a specific implementation of a utility such as JDBC FastExport.
• Utility classification and query characteristic classification are mutually exclusive. If you use one, the other option is not available.
• You can have one utility classification per rule.
Adding Utility Classification Type
You can classify filters, throttles, workloads, or sessions by utilities such as FastLoad or
MultiLoad.
1
2
3
4
5
From the Filters, Throttles, Workloads, or Sessions (click the Utility Limits or Utility
Sessions tab) button, select the name of an item or create one.
Click the Classification tab.
Do one of the following:
• From the Add Classification Criteria list, click Utility, and click Add.
• Select an existing utility criteria.
Select any combination of FastLoad, FastExport, MultiLoad, and Archive/Restore utilities. Select a utility type, such as FastLoad or FastExport, or a specific version, such as JDBC FastLoad or Stand Alone MultiLoad.
Click OK.
276 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Deleting a Classification Type
You can delete classification criteria from filters, throttles, workloads, query sessions, utility sessions, or utility limits at any time.
1
2
3
4
5
From the ruleset toolbar, click Filters, Throttles, Workloads, or Sessions (click the
Query Sessions, Utility Limits, or Utility Sessions tab).
In Name, select existing item.
Click the Classification tab.
Mouse over an existing classification criteria, and click . If you are deleting a utility limit criterion, clear the utilities you want to delete from the classification.
Click Save.
About Ruleset Sessions
This topic describes the session limit information you can specify when creating and editing rulesets. In Release 12.0 and Release 13.0, session throttles are configured on the Throttles tab. The Sessions view appears after you click the Sessions button on the ruleset toolbar and has the following tabs:
Query Sessions
Limits on the number of query sessions that can be logged on at one time. You can create, enable, clone, and delete query sessions. (Release 13.10 and later)
Query Sessions by State
Limits on the number of query sessions for each state. The default session limit for a state is listed, along with each state you have created and its assigned, state-specific session limit. (Release 13.10 and later)
Utility Limits
Limits on the number of utilities that can run at the same time. You can create, enable, clone, and delete utility limits.
Utility Limits by State
Limits on the number of utilities for each utility limit rule in each state. The default utility limit for a state is listed, along with each state you have created and its assigned, state-specific utility limit.
Utility Sessions
Limits on the number of sessions a specific utility can use. You can create, enable, clone, and delete utility sessions. (Release 13.10 and later)
Utility Sessions Evaluation Order
Precedence, from highest to lowest, of utility session rules. Evaluation order determines the rule in which the utility job is placed if a utility job matches more than one utility session rule. (Release 13.10 and later)
Teradata Viewpoint User Guide, Release 13.11
277
Chapter 28 Workload Designer
278
Creating a Query Session
You can create a query session in Release 13.10 or later. In Release 12.0 and Release 13.0, use a throttle. A query session limits the number of sessions that can be logged on at one time.
After the query session is created, additional controls in the Query Sessions tab allow you to clone, delete, and enable or disable the query session. View all created query sessions on the
Query Sessions by State tab.
1
2
3
4
5
6
7
8
9
10
11
12
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Click the Query Sessions tab.
Click Create Query Session.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select a Rule Type from the list:
• Select Collective if you want all users that meet the classification criteria treated as a group, with the group allowed a maximum number of queries.
• Select Individual if you want to apply limits to each user individually.
• Select Member if you want accounts or profiles that represent user groups used as the classification criteria for the rule. Limits are placed on each individual in the group, and no limit is placed on the account or group.
Click Save.
Click the Classification tab.
Add classification criteria.
Click the State Specific Settings tab.
Set state-specific session limits.
Setting Classification for Query Sessions
Add classification settings to existing query sessions or when creating a query session.
Classification options determine which sessions match a rule.
1
2
3
4
5
6
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Do one of the following in the Query Sessions tab:
• In Name, select an existing query session.
• Click Create Query Session, enter a name and optional description (up to 80 characters long), select a rule type, and click Save.
Click the Classification tab.
Click Add Criteria.
In Source Type, click and select one of the following:
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
7
8
• Account Name. The Teradata Database account name.
• Account String. The Teradata Database account identification string.
•
Profile. The Teradata Database profile name.
• Application. The application on the network client.
• Client IP Address. The IP address of the network client.
•
Client ID. The logon name on the network client.
Do at least one of the following:
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list.
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Click OK.
Setting State Specific Values for Query Sessions
Add state-specific settings to existing query sessions or when creating a query session.
You can override the default by specifying session limits on a per-state basis. For example, you might want to have session limits during high-traffic states and no session limits during low-traffic states. View all created query sessions on the Query Sessions By State tab.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
The Query Sessions tab appears, listing all existing query sessions and their attributes.
Do one of the following in the Query Sessions tab:
• In Name, select an existing query session.
• Select Create Query Session, enter a name and optional description (up to 80 characters long), select a rule type, and click Save.
Select State Specific Settings.
Click a state in the state matrix to set a query session limit for that state.
Select Create State Specific Settings.
Select Unlimited, or enter a session limit in the box.
Click OK.
Your selection is applied to each cell having that state, and overrides the setting specified in Default Settings.
[Optional] Change the default setting by selecting Unlimited, or enter a session limit number in the box.
Click Save.
Teradata Viewpoint User Guide, Release 13.11
279
Chapter 28 Workload Designer
280
Creating a Utility Limit
A utility limit determines the number and type of utility jobs that can be run at one time.
After the utility limit is created, additional controls in the Utility Limits tab allow you to clone, delete, and enable or disable the utility limit.
1
2
3
4
5
6
7
8
9
10
11
12
13
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Click the Utility Limits tab.
Click Create Utility Limit.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select Reject utilities that exceed limit to abort any utilities that exceed the limit. If you do not select this option, utilities are delayed.
(Release 12.0 only.)
Click Save.
Click the Classification tab.
Select the utilities to which the limit should be applied.
When a utility limit is created, several utilities are selected by default. The default utilities can be cleared.
Click Save.
Click the State Specific Settings tab.
Click Save.
Setting Classification for Utility Limits
Add classification settings to existing utility limits or when creating a utility limit.
Classification options determine which sessions match a rule.
1
2
3
4
5
6
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Click the Utility Limits tab.
Do one of the following:
• In Name, select an existing utility limit.
• Click Create Utility Limit, enter a name and optional description (up to 80 characters long), and click Save.
Click the Classification tab.
Select any combination of FastLoad, FastExport, MultiLoad, and Archive/Restore utilities. Select a utility type such as FastLoad or FastExport, or a specific version, such as .NET FastLoad or Stand-Alone MultiLoad. (In Release 12.0 and Release 13.0, specific versions are not available, and only one utility type can be selected.)
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
7
When a utility limit is created, several utilities are selected by default. The default utilities can be cleared.
Click Save.
Setting State-Specific Job Limits for Utility Limits
Set state-specific job limits for existing utility limits or when creating a utility limit.
You can override the default by setting job limits on a per-state basis. For example, you might want to raise the job limit during a low-traffic state, and lower the job limit during a high-traffic state.
1
2
3
4
5
6
7
8
9
10
11
12
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Select Utility Limits, and do one of the following:
• In Name, select a utility limit.
• Select Create Utility Limit, enter a name and optional description (up to 80 characters long), and click Save.
Select the State Specific Settings tab.
Click a state in the state matrix to select it.
Select Create State Specific Settings.
Select Job Limit.
Enter a number in the box.
Select Delay or Reject.
(Release 13.0 or later.)
Click OK.
Your selection is applied to each cell having that state, and overrides the settings specified in Default Values.
[Optional] To change the default setting:
a
Select Job Limit.
Your selection is applied to each cell having that state, and overrides the settings specified in Default Values.
b
Enter a number in the box.
c
Select Delay or Reject. (Release 13.0 or later.)
Click Save.
Creating a Utility Session
You can create a utility session in Release 13.10 or later. A utility session controls the number of sessions that are allowed to be logged on to each utility at one time. After the utility session is created, additional controls in the Utility Sessions tab allow you to clone, delete, and enable or disable the utility session.
Teradata Viewpoint User Guide, Release 13.11
281
Chapter 28 Workload Designer
282
1
2
3
4
5
6
7
8
9
10
11
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Click the Utility Sessions tab.
Click Create Utility Session.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select the utilities to which this session limit applies.
From the list, select the Data Size.
In Max Sessions, enter the maximum number of sessions that are allowed to be logged on to each of the selected utilities at one time.
Click Save.
Click the Classification tab to add and save classification criteria.
Setting Classification for Utility Sessions
Add classification settings to existing utility sessions or when creating a utility session.
Classification options determine which sessions match a rule.
1
Edit or create a ruleset.
2
3
4
From the ruleset toolbar, click Sessions.
Click the Utility Sessions tab.
Do one of the following:
• In Name, select an existing utility session.
• Click Create Utility Session, enter a name, select the utilities to which the session limit applies, select a Data Size, enter a Max Sessions number, and click Save.
5
6
7
Click the Classification tab.
In the Add Classification Criteria list, select Request Source or Query Band.
Click Add.
8
9
Specify options based on the classification criteria you selected. If you selected Request
Source, do at least one of the following:
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list.
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Click OK.
Setting Evaluation Order for Utility Sessions
You can create utility sessions and set evaluation order in Release 13.10 or later. If a utility job matches more than one utility session rule, evaluation order determines the rule in which the utility job is placed. The rule in the highest position on the Utility Sessions Evaluation
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
Order tab is applied. You cannot change the order of the system rules located at the bottom of the list.
1
2
3
4
5
Edit or create a ruleset.
From the ruleset toolbar, click Sessions.
Click the Utility Sessions Evaluation Order tab.
Drag rules to reorder the list of utility rules.
Click Save.
About Ruleset Filters
A filter rejects or limits a query before the query starts running. You can specify filters when creating or editing a ruleset. Add criteria to each filter that identifies the queries to which the filter should be applied. When creating or editing a ruleset, filters are specified by clicking
Filters in the ruleset toolbar. Any time after creating a filter, you can specify the states to which you want the filter applied.
Following are examples of using filters:
• Create a filter that prohibits a specific user from running a query with an estimated processing time of longer than 15 minutes.
• Create a filter to limit all members of a specific department that runs large reports from accessing the database during peak work hours.
Creating Filters
A filter rejects or limits user queries.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click Filters.
Click Create Filter.
Enter a name.
[Optional] Enter a description (up to 80 characters long).
[Optional] Select Global Rule - applies to all requests to apply global filters to all objects and, as a result, to all logon and query requests that meet the criteria of the rule.
If Global Rule - applies to all requests is selected, the request source and target classification types are not available. If request source or target classification type is used, the Global Rule - applies to all requests option is not available.
(Release 12.0 and Release 13.0 only.)
[Optional] Select Warning Only to have a warning message for the filter logged by the database. (Queries still run.)
Click Save.
Click the Classification tab.
Add and save classification criteria.
Teradata Viewpoint User Guide, Release 13.11
283
Chapter 28 Workload Designer
11
12
13
14
Include classification items in a filter to reject those items. For example, create a filter and add classification criteria to reject all queries from the finance department.
Click the State Specific Settings tab.
Enable or disable the filter for specific states.
Click OK.
Click Save.
Filter Classification Criteria
Some filters for query characteristics classification criteria are only available in Release 13.10.
Filter Options
Statement Type
AMP Limits
Rel 12.0
available
Step Row Count
Final Row Count available available
Estimated Processing
Time Rel
Minimum Step Time available
Join Type
Full Table Scan
AND or OR option to establish how multiple criteria are joined available available available
Rel 13.0
available available available available available available available
Rel 13.10
available available available available available available available available
Setting Classification for Filters
Filters limit or reject user queries. Add classification settings to existing filters or when creating a filter. Classification options determine which queries or sessions match a rule.
When using the query characteristics classification criteria with filters, consider the following:
• In Release 13.0 and earlier, select the AND or OR option to establish how multiple criteria are joined.
• In Release 13.0 and earlier, the Statement Type, AMP Limits, and Estimated Processing
Time criteria are available.
• In Release 13.10 or later, a Minimum option is available when using Step Row Count criteria.
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click Filters.
Do one of the following:
• In Name, select an existing filter.
284 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
4
5
6
7
8
9
• Click Create Filter, enter a name and optional description (up to 80 characters long), and click Save.
Click the Classification tab.
In the Add Classification Criteria list select Request Source, Target, Query
Characteristics, Query Band, or Utility.
Click Add.
Specify options based on the classification criteria you selected.
If you selected Request Source or Target, do at least one of the following:
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include or
Exclude buttons to add the match string to a list.
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Click OK.
Setting State-Specific Values for Filters
Add state-specific settings to existing filters or when creating a filter.
You can override the default by enabling or disabling the filter on a per-state basis. For example, you may want to leave a filter enabled under all circumstances except when a specific state occurs.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click Filters.
The Filters tab appears, listing all existing filters and their attributes.
Do one of the following in the Filters tab:
• In Name, select an existing filter.
• Select Create Filter, enter a name and optional description (up to 80 characters long), and click Save.
Select State Specific Settings.
Click a state in the state matrix to enable or disable the filter for that state.
Select Create State Specific Setting.
Select Filter is Enabled or Filter is Disabled.
Click OK.
Your selection is applied to each cell having that state, and overrides the settings specified in Default Settings.
[Optional] Change the default setting by selecting Filter is Enabled or Filter is
Disabled.
Click Save.
Teradata Viewpoint User Guide, Release 13.11
285
Chapter 28 Workload Designer
About Ruleset Throttles
A throttle limits the number of user queries that can be active at one time. When creating or editing a ruleset, you can specify throttles. Throttles are different from filters, which reject queries. After the throttle limit is reached, the workload adds new queries to the delay queue.
The following are examples of using throttles:
• Create a throttle that limits a specific user to running no more than 10 queries at a time.
• Create a throttle that limits a specific department to no more than 4 simultaneous queries.
When creating or editing a ruleset, throttles are specified by clicking Throttles in the ruleset toolbar.
Anytime after creating a ruleset throttle, you can specify the throttle limits for each state.
Creating Throttles
Throttles limit the number of user queries that can be active at one time. This is different from filters, which reject queries and prevent them from running at all.
1
2
3
4
5
6
7
8
9
10
11
12
13
Edit or create a ruleset.
From the ruleset toolbar, click Throttles.
Click Create Throttle.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select a Rule Type:
• Select Collective if you want all users that meets the classification criteria treated as a group, with the group allowed a maximum number of queries.
• Select Individual if you want to apply limits to each user individually.
• Select Member if you want accounts or profiles that represent user groups used as the classification criteria for the rule. Limits are placed on individuals in the group and no limit is placed on the account or group.
• Select Global if you want all queries placed in a single queue. If Global is selected, the request source and target classification types are not available. If request source or target classification type is used, the Global Rule option is not available.(Release 12.0
and Release 13.0.)
[Optional] Select Disable Manual Release or Abort to prevent Teradata Database
Administrators from aborting or releasing throttled queries in the queue.
[Optional] Select Reject throttled queries to abort throttled queries. (Release 12.0.)
Click Save.
Click the Classification tab.
Add and save classification criteria.
Click the State Specific Settings tab.
Define query limits for specific states.
286 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
14
15
Click OK.
Click Save.
Throttle Classification Criteria
Some throttles for query characteristics classification criteria are only available in Release 13.10.
Filter Options
Statement Type
Rel 12.0
available
AMP Limits available
Any Step Time Exceeds available
Step Row Count
Final Row Count
Estimated Processing
Time
Minimum Step Time
Join Type
Full Table Scan multiple criteria joined by AND
Rel 13.0
available available available
Rel 13.10
available available available available available available available available available
Setting Classification for Throttles
Throttles limit the number of user queries that can run at the same time. Add classification settings to existing throttles or when creating a throttle. Classification options determine which queries match a rule.
When using the query characteristics classification criteria with throttles in Release 13.0 or earlier, the Statement Type, AMP Limits, and Any Step Time Exceeds criteria are available.
1
2
3
4
5
6
7
8
Edit or create a ruleset.
From the ruleset toolbar, click Throttles.
In the Throttles tab, do one of the following:
• In Name or Workload Name, select a throttle.
• Click Create Throttle, enter a name and optional description (up to 80 characters long), select a rule type, and click Save.
Click the Classification tab.
In the Add Classification Criteria list, select Request Source, Target, Query
Characteristics, Query Band, or Utility.
Click Add.
Specify options based on the classification criteria you selected.
If you selected the Request Source or Target criteria, do at least one of the following:
Teradata Viewpoint User Guide, Release 13.11
287
Chapter 28 Workload Designer
9
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list.
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Click OK.
Setting State-Specific Values for Throttles
Set state-specific query limits for existing system throttles or when creating a throttle.
You can override the default by setting query limits on a per-state basis. For example, you may want to raise the query limit of a system throttle during a low-traffic state, and lower the query limit during a high-traffic state.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click Throttles.
The Throttles tab appears, listing all existing throttles and their attributes.
Do one of the following:
• Select an existing system throttle from the System Throttles list.
• Select Create Throttle, enter a name and optional description (up to 80 characters long), select a rule type, and click Save.
Select State Specific Settings.
In When Query Limit Exceeded, select Delay or Reject. (Release 12.0.)
Click a state in the state matrix to set the throttle query limit for that state.
In Edit [state name] Settings, select Create State Specific Settings.
Note: If you select the Use Default Settings option, the settings at the bottom of the
Throttles tab apply to the state.
Do one of the following:
• Next to Query Limit, select Unlimited to place no limits on the number of queries for this throttle.
• Enter a query limit number in the box, and select either Delay or Reject. (The Delay and Reject options are available in Release 13.0 or later.)
• Next to Session Limit, select Unlimited or enter a session-limit number in the box.
(Release 12.0 and Release 13.0.)
Click OK.
Your selection is applied to each cell having that state, and overrides the setting specified in Default Settings.
[Optional] Change the Default Settings by selecting Unlimited, or enter a query limit number in the box and select Delay or Reject. (The Delay and Reject options are available in Release 13.0 or later.)
288 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
11
12
[Optional] Change the Session Limit by selecting Unlimited or entering a session limit number in the box. (Release 12.0 and Release 13.0.)
Click Save.
About Ruleset Workloads
A ruleset workload is a group of queries that share characteristics so that a set of workload management controls can be applied to the group. A workload has working values and defining characteristics that are evaluated during the classification phase of system management. High-quality workload management can improve response times and ensure more consistent response times for critical work.
The Workloads tab lists workload names, types, throttle limits, and enabled status. For each workload, you can specify one or more of the following:
• Classification criteria which determine if a query is assigned to the workload.
• Throttles which limit the number of concurrent active queries that can run.
• Service level goals which specify a goal for workload query performance.
• Exceptions which monitor queries and take specified actions if a query exceeds exception criteria while executing.
You can create up to 250 workloads. The following requests are examples of workloads:
• Batch jobs further subdivided by region or organization for reporting.
• Weekly or monthly reports that follow the calendar or a regular schedule.
• Jobs that are always critical whenever they occur.
Queries not classified into a specific workload are placed into the default workload named
WD-Default. The WD-Default workload cannot be deleted or disabled.
Creating a Workload
You can group queries that share characteristics into a workload so that a set of workload management controls can be applied to the group.
1
2
3
4
5
6
7
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click Create Workload.
In the General tab, enter a name.
[Optional] Enter a description up to 80 characters long.
Select an Enforcement Priority (the type of queries you expect to run in this workload):
• Tactical queries are short and high-priority.
• Priority queries are longer-running than tactical queries and higher priority.
• Normal queries are the customary queries running on the system.
• Background queries run when the system is not busy.
Click Save.
Teradata Viewpoint User Guide, Release 13.11
289
Chapter 28 Workload Designer
8
9
10
11
12
13
14
[Optional] Click the Classification tab.
[Optional] Determine if you want incoming queries classified into this workload. For example, select Target to have queries from a specific database sent to this workload.
[Optional] Click the Throttles tab.
[Optional] Determine if you want to set a throttle just for this workload. Any query that is classified into this workload is subject to this throttle.
[Optional] Click the Service Level Goals tab, and do one of the following:
• Select Response Time Goal, and enter a time in seconds for how quickly you want the queries to run. For example, type 4 if you want all queries to finish within 4 seconds. You can also set a service percentage. For example, type 90 if you expect 90% of queries to meet the time goal.
• Select Throughput Goal, and type a throughput number per hour. For example, type
200 if you expect 200 queries to be processed per hour.
• Select No Goal if you want no service-level goal for this workload.
[Optional] Click the Exceptions tab. Determine if you want an exception triggered for this workload. For example, if a query in this workload is taking too long, create an exception to move the query to a different workload.
Click Save.
Setting Classification for Workloads
Classification options determine which queries go into a workload. Add classification settings to existing workloads or when creating a workload. Classification options cannot be added to the default workload, WD-Default.
1
2
3
4
5
6
7
8
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Do one of the following:
• In Name, select an existing workload.
• Click Create Workload, enter a name and optional description (up to 80 characters long), select an Enforcement Priority, and click Save.
Click the Classification tab.
Select classification criteria from the list.
The classification types are Request Source, Target, Query Characteristics, Query
Band, and Utility.
Click Add.
Specify options based on the classification criteria you selected.
If you selected Request Source or Target, do at least one of the following:
• In Match String, enter a string. A match string can contain ? to match exactly one character or an * to match zero or more characters. Use the Match String Include and Exclude buttons to add the match string to a list.
290 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
9
• Select Items from the list and use the Items Include and Exclude buttons to create your classification criteria.
Click OK.
Setting State-Specific Throttles for Workloads
Set state-specific throttle query limits for existing workloads or when creating a workload. If you select the Use Default Settings option, the settings at the bottom of the Throttles tab apply to the state.
1
2
3
4
5
6
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Do one of the following:
• In Name, select an existing workload.
• Click Create Workload, enter a name and optional description (up to 80 characters long), select an Enforcement Priority, and click Save.
Click the Throttles tab.
Click on a state in the state matrix.
In Edit [state name] Settings, select Create State Specific Settings.
7
Do one of the following:
• Select Unlimited to prevent limits on the number of queries for this workload.
Teradata Viewpoint User Guide, Release 13.11
291
Chapter 28 Workload Designer
8
9
10
• Enter a query limit number in the box, and select either Delay or Reject. (The Delay option is available in Release 13.0 and later.)
Click OK.
Your selection is applied to each cell having that state and overrides the setting specified in Default Settings.
[Optional] Change the default setting by selecting Unlimited or entering a number in the box so queries over the limit are rejected. (In Release 13.0 and later, Delay and
Reject options are available.)
Click Save.
Setting Service-Level Goals
Service level goals (SLGs) help determine if workload management is meeting expectations.
You can state service-level goals in throughput or response time with a service percentage.
1
2
3
4
5
6
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Do one of the following:
• In Name, select an existing workload.
• Click Create Workload. Enter a name, optional description (up to 80 characters long), and enforcement priority. Click Save.
Click the Service-Level Goals tab.
Do one of the following:
• Select Response Time Goal, enter a time in seconds, and set a service percentage. For example, if you want 90% of queries to finish within 4 seconds, set a response time of
4 and a service percent of 90. You can have different goals in different planned environments. For example, you can set a response time of 4 seconds for a daytime environment and 12 seconds for a nighttime environment. Metrics graphs show response times.
• Select Throughput Goal and enter a throughput number per hour.
• Select No Goal if you want no service-level goal. This option may be appropriate, for example, for very low priority background workloads.
Click Save.
Defining a Workload Exception
A workload exception is one or more events that, when all occur at the same time (or continuously for a specified period of time) cause a defined action to occur. You can add multiple criteria to an exception. If you add more than one criteria for an exception, all criteria must be satisfied for the exception to be triggered.
Note: Some exception criteria give you the option to add qualification time. Qualification time is the amount of time the criteria must take place to be recognized. Setting qualification time prevents very short incidents of a criteria from being acknowledged.
292 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Do one of the following:
• In Name, select an existing workload.
• Click Create Workload, enter a name, optional description (up to 80 characters long), and enforcement priority, and click Save.
Click the Exceptions tab.
Note: In Release 12.0 and Release 13.0, local exceptions are displayed separately from shared exceptions on the Exceptions tab.
Click Create Exception.
Enter a name.
[Optional] Enter a description up to 80 characters long.
In Criteria, select an option from the list.
Click Add.
Specify the additional required information.
In Actions, select one of the following actions to be performed when the specified criteria occur:
• Select Notification Only to send notification, and take no other action.
• Select Abort to stop the query.
• Select Abort Selects Only to stop the query only if it is a SELECT.
• Select Change Workload to to change the query workload to the workload you select from the list.
[Optional] In Notifications, select any of the following options:
• Select Alert to specify the Teradata Alerts to trigger at the start of the event. Select an alert name from the list.
• Select Run Program to specify the Teradata Alerts registered programs to trigger at the start of the event. Select a program from the list.
• Select Post to QTable to post the text you specify to the system queue table when the event starts. (This option is not integrated with Teradata Alerts.)
Click OK.
If there is more than one local exception assigned to the Change Workload action, use the Exception Precedence tab to set the priority of the exceptions. (Release 12.0 and
Release 13.0.)
Viewing Workload Service Level Goal Summary
A service level goal for workload query performance is set in either throughput or response time with a service percentage. You can view a summary of service level goals that have been set for individual workloads.
Teradata Viewpoint User Guide, Release 13.11
293
Chapter 28 Workload Designer
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the SLG Summary tab.
Workloads are listed, along with their service level goals for each planned environment.
About Evaluation Order
Evaluation order determines the placement of queries into workloads and utility jobs into rules. Setting evaluation order is useful when you have created several workloads or utility sessions.
For workloads, if classification criteria can sort the same query into different workloads, evaluation order determines the workload in which the query is placed. For example, account finance users are classified into one workload and users in a specific geographical location are classified into a different workload. If an individual is a member of both groups, the workload highest on the evaluation order list is the workload in which the individual is sorted.
For utility sessions, if a utility job matches more than one utility session rule, evaluation order determines the rule in which the utility job is placed.
Setting Evaluation Order for Workloads
If classification criteria can sort the same query into different workloads, evaluation order determines the workload in which the query is placed. The workload in the highest position on the Evaluation Order tab is applied.
1
2
3
4
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Evaluation Order tab.
Select a Planned Environment from the list.
294
5
In the list of workloads, do one of the following to order the list and place higher priority workloads at the top:
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
6
• Drag workload names to reorder.
• Click in a number box, highlight the existing number, and enter a new number.
Click Save.
Setting Enforcement Priority for Workloads
Enforcement priority establishes the type of queries you expect to run in a specific workload.
1
2
3
4
5
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Do one of the following in the Workloads tab:
• In Name, select an existing workload.
• Click Create Workload, and enter a name.
Select an Enforcement Priority:
• Tactical queries are short and high-priority.
• Priority queries are longer-running than tactical queries and high-priority.
• Normal queries are the regular queries running on the system.
• Background queries run when the system is not busy.
Click Save.
Adjusting Priority Distribution for Workloads
On the Priority Distribution tab, each cell in the diagram is an allocation group. The size of each cell is proportional to the relative weight. Each column in the diagram is a resource partition. The width of each column is proportional to the relative weight of the resource partition. Drag cell borders to dynamically change the amount of CPU (as a percentage) allocation groups are given. Drag column borders to change the amount of CPU available to resource partitions.
Note: If you modify the base state on the Priority Distribution tab, the changes are applied to other states as long as no change has been made to any of the other states. Once any of the states are changed, the only values that are applied are the changes to system values.
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Priority Distribution tab.
Teradata Viewpoint User Guide, Release 13.11
295
Chapter 28 Workload Designer
296
4
5
6
7
8
In State, select a state from the list.
For resource partitions, do any of the following:
• Mouse over any resource partition name to obtain information about the partition.
• Click on a resource partition name, and drag the column resize bar. This resizes the column and reallocates space between the resource partitions.
• Click on a resource partition name, and click . Enter a system percentage allocation and, optionally, a CPU percentage limit. If you select the Locked option, this column becomes fixed in the diagram, and other columns can be dynamically resized around this column.
• Click on a resource partition name and click or to unlock or lock the column.
For allocation groups, do any of the following:
• Mouse over any allocation group to obtain information about the group.
• Click in a cell, and drag the cell resize bar. This resizes the cell and reallocates space within a resource partition.
• Click in a cell, and click . Enter a system percentage allocation and, optionally, a
CPU percentage limit. If you select the Locked option, this cell becomes fixed in the diagram and other cells can be dynamically resized around this cell.
• Click in a cell and click or to unlock or lock the cell.
[Optional] Select System Values, and enter a CPU limit to set the maximum percentage of system CPU that Teradata Database can use.
[Optional] Select Reserved AWTs (AMP Worker Task), and enter a number to set the maximum number of AWTs that are reserved for tactical workloads.
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
9
10
[Optional] Select Max Expedited AWTs, and enter a number to set the maximum number of AWTs that are created on each node.
Click Save.
Mapping Console Utilities to Workloads
Map console utilities to workloads so utilities can be organized and prioritized.
1
2
3
4
5
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Console Utilities tab.
In Console Utility to Workload Mapping, use the lists to set a workload name for each console utility.
The lists contain the predefined workload, WD-Default, and any workloads you have created. WD-Default is used when no other workload is specified.
Click Save.
About Performance Groups
A performance group is a collection of parameters used to control and prioritize resource allocation for a particular set of database sessions. Performance groups belong to a resource partition, a set of prioritized performance groups related by their users' associations.
Resource partitions have an assigned weight that determines the proportion of resources available to that partition relative to other partitions.
On the Console Utilities tab, the mapping of performance groups to workloads only applies to console utilities that are not included in the Console Utility to Workload Mapping section.
Creating a Performance Group
Create a performance group to determine which allocation group manages a query.
Performance groups belong to a resource partition, a collection of prioritized performance groups related by their users' associations.
1
2
3
4
5
6
7
8
9
10
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Console Utilities tab.
Click Create Performance Group Mapping.
Mouse over New Performance Group, and click .
Enter a name.
Click outside the name.
In Workload Name, select a workload from the list for the performance group.
Click Save.
[Optional] To delete a performance group, mouse over the group name, and click .
Teradata Viewpoint User Guide, Release 13.11
297
Chapter 28 Workload Designer
298
Mapping Performance Groups to Workloads
Map performance groups to workloads so the groups can be prioritized along with other work.
1
2
3
4
5
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Console Utilities tab.
In Performance Group to Workload Mapping, use the lists to set a Workload Name for each Performance Group.
The lists contain the predefined workload, WD-Default, and any workloads you have created. WD-Default is used when no other workload is specified.
Click Save.
About Allocation Groups
An allocation group can limit the total amount of CPU used by sessions under its control. An allocation group has an assigned weight that is compared to other allocation group weights.
The four default allocation groups are tactical, priority, normal, and background. The default allocation groups and any groups you create are assigned to resource partitions.
Note: The allocation groups named L, M, H, and R stand for Low, Medium, High, and Rush.
Do not use or modify the allocation groups named L, M, H, and R. The groups represent levels of priority and are used by the system for console utilities.
By default, individual workloads are associated with a specific allocation group based on the enforcement priority assigned to the workload. You can change the default mapping using the diagram on the Workload Mapping tab. The diagram enables you to reassign workloads to allocation groups and reassign allocation groups to resource partitions. Allocation groups that contain workloads cannot be deleted.
Carefully consider the allocation groups you drag in and out of the tactical resource partition because this can have a significant effect on the amount of CPU available to the queries running in the allocation group. The tactical resource partition normally has a much higher weight than other resource partitions. If you move an allocation group containing long, resource-intensive queries into the tactical resource partition, those queries can consume a large amount of the available CPU, impacting queries running in other resource partitions. If you move a tactical allocation group out of the tactical resource partition, the queries running in the tactical allocation group are assigned a lower priority and may not meet their response-time goals.
Creating an Allocation Group
Create allocation groups using the Workload Mapping tab.
1
2
3
4
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Workload Mapping tab.
Mouse over the name of the resource partition, such as Tactical or Standard, to which you want to add an allocation group, and click .
Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
5
6
7
8
9
Enter a name.
Click outside the name.
[Optional] To move the allocation group to a different resource partition, mouse over the name of the group, and drag the group to a different column.
[Optional] To change the name of the allocation group, mouse over the name of the group, and click .
Click Save.
Mapping Allocation Groups to Workloads
If you have created workloads, you can view the Workload Mapping tab to see how the workloads map to allocation groups and resource partitions. Allocation groups and resource partitions determine how much CPU is available to the queries that are running in the listed workloads. The allocation groups named L (Low), M (Medium), H (High), R (Rush) represent levels of priority. These groups are used by the system for console utilities and cannot be deleted.
1
2
3
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Workload Mapping tab.
4
Do any of the following:
• Create allocation groups by mousing over a resource partition name, such as Tactical or Standard, and clicking .
• Create a resource partition by mousing over Resource Partitions and clicking .
Teradata Viewpoint User Guide, Release 13.11
299
Chapter 28 Workload Designer
5
• Drag an allocation group to a different resource partition.
• Drag a workload to a different allocation group of same priority in any resource partition.
• Rename a resource partition by mousing over a name and clicking .
• Delete an empty allocation group by mousing over a name and clicking .
• Delete an empty resource partition by mousing over a name and clicking .
Click Save.
Deleting An Allocation Group
You cannot delete the L (Low), M (Medium), H (High), R (Rush) allocation groups or allocation groups that contain workloads.
1
2
3
4
5
Edit or create a ruleset.
From the ruleset toolbar, click Workloads.
Click the Workload Mapping tab.
Mouse over the name of an allocation group, and click .
Click Save.
About Ruleset Exceptions
An exception is a collection of one or more events that, when all occur at the same time (or continuously for a specified period of time), cause a defined action to occur. When creating or editing a ruleset, you can specify a set of exception criteria and exception actions to apply to workloads for each planned environment.
The tabs on the ruleset Exceptions view display all exception information for a ruleset from different perspectives:
By Planned Environment
Exceptions organized by planned environment. After selecting a planned environment, all exceptions are listed alphabetically by name. All workloads are listed in workload evaluation order. Select and clear the workload check boxes to enable or disable the use of individual exceptions for individual workloads. If the All-
WD check box is selected, all check boxes in the row are selected and cannot be individually controlled.
By Workload
Exceptions organized by workloads. After selecting a workload, all exceptions are listed by name. All planned environments are listed by name and by how they are arranged in the state matrix. Select and clear the planned environment check boxes to enable or disable individual exceptions for individual planned environments.
Within a workload, if an exception is enabled for All-WD in a specific planned environment, the check box is selected and disabled. (You can change the exception setting on the Planned Environment tab.) In Release 12.0 and Release 13.0 local exceptions are displayed separately from shared exceptions.
300 Teradata Viewpoint User Guide, Release 13.11
Chapter 28 Workload Designer
By Exception
List of all exceptions created for planned environments and workloads. After selecting an exception, all workloads are listed in workload evaluation order. All planned environments are listed by name. Select and clear the planned environment check boxes to enable or disable individual workloads for individual planned environments. If the All-WD check box is selected, all check boxes in the column are selected and cannot be individually controlled.
Defining an Exception
An exception is defined by a collection of one or more threshold events that, when all occur at the same time and for a specified period of time, cause a defined action to be taken. If you add more than one criteria, all criteria must be satisfied in order for the exception to be triggered. Only one qualification time can be specified even if multiple qualified criteria are added. If qualified criteria are specified, the exception is triggered only if all of the qualified criteria are satisfied for the duration of the qualification time. After an exception is created, additional controls in the tabs of the Exceptions view allow you to manage the exception.
1
2
3
4
5
6
7
8
9
10
11
12
Edit or create a ruleset.
From the ruleset toolbar, click Exceptions.
From any of the tabs in the Exception view, click Create Exception.
Enter a name.
[Optional] Enter a description up to 80 characters long.
Select an option from the Exception Criteria list, and click Add.
Enter a value for the criteria.
For qualified criteria (marked with *), enter a value in Qualification Time.
Select one of the following actions to be performed when the specified criteria are satisfied:
• Notification Only sends a notification only. No other action is performed.
• Abort stops the query.
•
Abort Selects Only stops the query only if it is a SELECT.
• Change Workload to changes the workload of the query to the specified workload.
Change Workload to is unavailable if the Blocked Time criterion or Elapsed Time criterion was selected.
[Optional] Select any of the following notifications:
•
Alert sends the selected alert.
• Run Program starts the selected program.
• Post To QTable posts the string entered in the box to the QTable. (In Release 12.0, you can select the Post to QTable option, but the text box is not available.)
Click OK.
If there is more than one shared exception assigned the Change Workload to action, use the Exception Precedence tab to set the priority of the exceptions.
Teradata Viewpoint User Guide, Release 13.11
301
Chapter 28 Workload Designer
(Release 12.0 and Release 13.0.)
About Tactical CPU per Node Exceptions
If you select the Tactical CPU per Node exception criteria, the following rules apply:
• The exception can be enabled for workloads that use the tactical enforcement priority.
Tactical workloads are in the top row of the Workload Mapping tab.
• The exception must have a Change Workload to action to another workload in the same resource partition column.
• The exception must have Tactical CPU per Node set to five cpu-seconds or less.
• The exception must have CPU Time set to a number greater than Tactical CPU per
Node.
• Using Tactical CPU per Node criteria disables the Blocked Time and Elapsed Time criteria for the exception.
Managing Exceptions
Ruleset exceptions are managed from the ruleset Exceptions view. In Release 12.0 and
Release 13.0, local exceptions are displayed separately from shared exceptions on the by
Workload tab.
1
2
3
4
5
6
7
8
9
Edit or create a ruleset.
From the ruleset toolbar, click Exceptions.
[Optional] Create exceptions by clicking Create Exception from any tab.
After exceptions have been created, click the by Planned Environment, by Workload, or
by Exception tab to manage exceptions.
Depending on the tab, select Workload or Exception from the list.
[Optional] Edit an exception from the by Planned Environment or by Workload tab by mousing over the exception and clicking .
[Optional] Delete an exception from the by Planned Environment or by Workload tab by mousing over the exception and clicking .
In the selected view, select or clear the appropriate check boxes to enable or disable the exceptions for the listed workloads, planned environments, or workload and planned environment combinations.
Click Save.
302 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 29
Workload Health
Workload Health
The Workload Health portlet displays workload health information for one system at a time. You choose which system, workloads, and metrics to monitor. Filter and Sort menus allow you to customize the information displayed. Data in the Workload Health portlet is refreshed every minute.
The Workload Health portlet displays workloads that:
• Have completed processing according to their Service Level Goals
• Have missed their Service Level Goals
• Are inactive
• Are disabled
• Have no defined Service Level Goals
The workload name HEALTH DETAILS view displays detailed metrics for an individual workload.
Workload Health Concepts
What is a Workload?
A workload is a group of queries that share characteristics so that a set of workload management controls can be applied to the group. A workload has working values and defining characteristics that are evaluated during the classification phase of system management.
Workload Examples
Name
DBA_GrandRapids
Characteristics
WD-Accounting-All Requests for the account strings for the entire accounting department. The characteristics are the accountIDs and region identifiers.
Requests submitted by the DBA Department in Grand Rapids. An IP address or user name could be specified as characteristics.
MULTILOAD_SMP Requests that execute the Teradata MultiLoad utility and single-AMP queries. These requests require fast response times.
What is a Service Level Goal?
A Service Level Goal (SLG) measures whether workload management is meeting defined criteria. An SLG can be expressed as a number or percentage. SLGs are defined for logging
Teradata Viewpoint User Guide, Release 13.11
303
Chapter 29 Workload Health and reporting purposes. There are no provisions in the Teradata Database to enforce these goals.
Not all workloads need an SLG. In general, it is good practice to establish SLGs for tactical workloads. However, low-priority, background workloads do not need an SLG.
What is an Enforcement Priority?
When a workload is created, it is assigned an enforcement priority (EP) and an allocation
group (AG). An EP is a ranking given to establish a processing order for the queries running within the workload. Each AG has only one EP, and an EP can have multiple AGs. Teradata
Database uses the following EPs:
Tactical
High-priority, critical queries with a short response-time requirement for single or few-AMP queries or all-AMP queries that consume less than 1 CPU second per node.
Priority
Important queries that might need extra resources.
Normal
Default enforcement priority for queries.
Background
Low-priority queries with no response-time requirements or Service Level Goals.
About the Workload Health View
Use the WORKLOAD HEALTH view to display the health status of one or more workloads.
Workload health is determined in relation to a Service Level Goal (SLG).
304
The following list describes the features in this view:
• System name in the portlet frame, color-coded to red if a workload has missed its SLG
• Active ruleset name (the ruleset currently enabled on the Teradata Database system)
• Workload names
• Workload health, presented using color, icons, and predefined states
Teradata Viewpoint User Guide, Release 13.11
Chapter 29 Workload Health
• Workload sort and filter capabilities
• Portlet rewind and share capabilities
Click a workload icon to go to the workload name HEALTH DETAILS view. The workload
name HEALTH DETAILS view is not available for workloads having a health state of
NO_DATA.
Workload Health States
Workload health is described using a set of icons and predefined states. Health states are listed in the default sort order.
Icon Text
Color
Red
State
MISSED_SLG
Description
Workload missed its SLG for a period of time, based on the enforcement priority:
• Tactical: last minute
• Priority: last 30 minutes
• Normal and Background: last 60 minutes
Blue
MET_SLG
Black
NO_SLG
Workload met its enforcement-priority Service Level
Goals.
This health state is determined by comparing the measured query Response Time with the expected
Response Time and Percent SLG metrics.
No Service Level Goals have been set for this workload.
Blue
NO_DATA
Blue
DISABLED
Gray
INACTIVE
No data is available for this workload.
The workload has been disabled.
The workload is inactive.
Selecting Filter Criteria
1
2
3
4
5
Click Filter.
The Filter Workloads dialog box appears.
[Optional] Select the All Enforcement Priorities check box or click to display the enforcement priority list and select at least one enforcement priority.
[Optional] Select the All Health States check box or click to display the health state list and select at least one health state.
[Optional] Select the Name check box and enter a workload name or name pattern.
Click to close the menu.
Setting Sort Criteria
1
Click Sort.
Teradata Viewpoint User Guide, Release 13.11
305
Chapter 29 Workload Health
2
The Sort menu appears.
To specify the sort criteria, do one of the following:
• Click Missed SLG to sort workloads into workload health state groups in alphabetical order by workload name.
• Click Name to display workloads in alphabetical order by workload name, regardless of workload health state.
About the Workload Health Details View
The workload name HEALTH DETAILS view displays metrics for a single workload. Use the PREFERENCES view to select the metrics. This view appears after you click the workload icon, name, or information balloon for a workload in the WORKLOAD HEALTH view. The workload name HEALTH DETAILS view is not available for workloads having a health state of NO_DATA.
306
Metrics are arranged in a two-column format:
• Metric names
• Sparkline graphs
As you mouse over the sparklines, information balloons display event information. Each sparkline ends with a percentage or a value, depending on the metric.
About Workload Metrics Graphs
Sparkline graphs illustrate workload performance metrics.
Solid lines represent the current data for their respective metric.
Teradata Viewpoint User Guide, Release 13.11
Chapter 29 Workload Health
On a single-sparkline graph (for example, the MAX RESPONSE metric), the solid line represents current data for the metric, and the grayed skyline represents the past averages for that same metric.
On a multi-sparkline graph, a metric can be expressed in more than one value. When you mouse over the sparkline, the color of the metric name, end data points, and information balloons match.
A dotted line in the graph is called the SLG reference line. Data points above the SLG reference line are color-coded in red.
Workload Metrics
Metric
Aborted Queries
Active Queries
Description
Number of aborted queries in the last 60 seconds.
Number of active queries in the last 60 seconds. This metric is a default for the Priority, Normal, and
Background EPs.
Arrivals Number of queries arriving in the last 60 seconds.
Historical data is not displayed for this metric. This is the default metric for all EPs.
Avg CPU-SEC per Query Average CPU seconds per completed queries is a default metric for the Tactical enforcement priority, providing throughput and processing information.
Avg Response & Delay
Time
Cumulative CPU
Default metric for all EPs.
Cumulative CPU seconds per amp. This is a default metric for the Priority, Normal, and Background EPs.
Type
Number
Number
Number
Number
Number
Number
Teradata Viewpoint User Guide, Release 13.11
307
Chapter 29 Workload Health
Metric
Delayed Queries
Exceptions
Description
Number of queries in the delay queue in the last 60 seconds. This metric is a default metric for Priority,
Normal, and Background EPs.
Number of queries classified as exceptions in the last
60 seconds.
Max CPU-SEC per Query Maximum CPU seconds used by completed queries is a default metric for the Tactical EP.
Maximum Response Time Default metric for the Tactical EP.
Queries Throttled
Rejected Queries
Throughput
Number of queries throttled in the workload in the last 60 seconds.
Number of rejected queries in the last 60 seconds.
Number of completed queries in the last 60 seconds.
This metric does not display historical data.
Type
Number
Number
Number
Number
Number
Number
Number
About the Preferences View
The PREFERENCES view allows you to customize the Workload Health portlet.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
System and Workloads
Choose a system and its workloads to monitor.
Metrics-Tactical
Select metrics for workloads in the Tactical enforcement priority.
Metrics-Priority
Select metrics for workloads in the Priority enforcement priority.
Metrics-Normal
Select metrics for workloads in the Normal enforcement priority.
Metrics-Background
Select metrics for workloads in the Background enforcement priority.
Trend Interval
Set the time period for calculating each trend interval.
Past Average
Choose the time period for which a historical average is calculated.
Selecting a System and Workloads to Monitor
Use the System and Workloads tab in the PREFERENCES view to select the system and which of its workloads to monitor.
308 Teradata Viewpoint User Guide, Release 13.11
Chapter 29 Workload Health
1
2
3
4
5
From the portlet frame, click to access the PREFERENCES view.
Click the System and Workloads tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a system from the list.
Select the check box for each workload you want to monitor.
6
7
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Enforcement Priority Metrics
For each enforcement priority group, you can select and order metrics for display from the
PREFERENCES view. Choose from a defined list of metrics.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab for an enforcement priority group.
The Preview pane uses sample data to show how metric rows are displayed in the summary view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do any of the following to change the way metric rows are displayed:
• Add a metric row. Drag a metric from the Select metrics for display list to the
Preview pane.
• Remove a metric row. Drag a metric from the Preview pane to the Metric Removal pane.
• Change metric row order. Drag a metric already in the Preview pane to a new location in the pane (up or down).
Changes to the metrics affect only the system currently selected, provided that the metrics are available on that system.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting the Workload Trend Interval
Use the Trend Interval tab in the PREFERENCES view to set the length of time used to plot sparkline metrics in the portlet details view.
1
2
From the portlet frame, click to access the PREFERENCES view.
Click the Trend Interval tab.
Teradata Viewpoint User Guide, Release 13.11
309
Chapter 29 Workload Health
3
4
5
6
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
In the dialog box, select an Interval.
The default is 24 hours.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Past Averages
Use the Past Averages tab in the PREFERENCES view to specify the number of weeks of data points used to calculate the average values displayed in metric graphs.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Past Averages tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter the number of weeks (1 to 99 weeks).
The default is 2 weeks.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
310 Teradata Viewpoint User Guide, Release 13.11
CHAPTER 30
Workload Monitor
Workload Monitor
The Workload Monitor portlet allows you to monitor workload activity, allocation group, and session data in Teradata Database.
The Workload Monitor portlet provides:
• Multiple summary and details views for presenting information
• A state matrix icon that displays the status of the Teradata Database system
• A choice of data sampling periods
• The ability to filter workloads and sort columns
Workload Monitor Concepts
What is a Workload?
A workload is a group of queries that share characteristics so that a set of workload management controls can be applied to the group. A workload has working values and defining characteristics that are evaluated during the classification phase of system management.
Workload Examples
Name Characteristics
WD-Accounting-All Requests for the account strings for the entire accounting department. The characteristics are the accountIDs and region identifiers.
DBA_GrandRapids Requests submitted by the DBA Department in Grand Rapids. An IP address or user name could be specified as characteristics.
MULTILOAD_SMP Requests that execute the Teradata MultiLoad utility and single-AMP queries. These requests require fast response times.
What is a Service Level Goal?
A Service Level Goal (SLG) measures whether workload management is meeting defined criteria. An SLG can be expressed as a number or percentage. SLGs are defined for logging and reporting purposes. There are no provisions in the Teradata Database to enforce these goals.
Not all workloads need an SLG. In general, it is good practice to establish SLGs for tactical workloads. However, low-priority, background workloads do not need an SLG.
Teradata Viewpoint User Guide, Release 13.11
311
Chapter 30 Workload Monitor
312
What is an Enforcement Priority?
When a workload is created, it is assigned an enforcement priority (EP) and an allocation
group (AG). An EP is a ranking given to establish a processing order for the queries running within the workload. Each AG has only one EP, and an EP can have multiple AGs. Teradata
Database uses the following EPs:
Tactical
High-priority, critical queries with a short response-time requirement for single or few-AMP queries or all-AMP queries that consume less than 1 CPU second per node.
Priority
Important queries that might need extra resources.
Normal
Default enforcement priority for queries.
Background
Low-priority queries with no response-time requirements or Service Level Goals.
Workload Metrics
Metric
Active Requests
Active Workloads
Average Response Time and Average Delay Time
Description
Number of active queries in the last sample period.
This is a default metric for the Priority, Normal, and
Background enforcement priorities.
Number of workloads that have queries classified into them over the user-defined time frame.
Average response time and the average delay time for all active requests. This is a default metric for all enforcement priorities.
Type
Number
Number
Number
Maximum Response Time Maximum response time for all active requests. This is a default metric for the Tactical enforcement priority.
Delayed Queries Number of queries in the delay queue in the last sample period. This metric is a default metric for
Priority, Normal, and Background enforcement priorities.
Average CPU-SEC per
Request
Maximum CPU-SEC per
Request
Average CPU time, in seconds, of workload requests.
This is a default metric for the Tactical enforcement priority.
Maximum CPU time, in seconds, of workload requests. This is a default metric for the Tactical enforcement priority.
Number
Number
Number
Number
Throughput Number of queries arriving and completing in the last
60 seconds. This metric does not display historical data. This is a default metric for all enforcement priorities.
Number
Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
Metric
Impact CPU-SEC
Cumulative CPU
Description
Maximum amount of CPU time used by a workload compared to the total CPU used by all workloads on the system.
Cumulative CPU seconds per amp. This is a default metric for the Priority, Normal, and Background enforcement priorities.
Type
Number
Number
What is an Exception?
A workload management exception occurs when a query executes outside of the system regulation that is currently in effect. The Workload Monitor portlet reports on exceptions for CPU or I/O skew, an excessive number of concurrent workloads, or queries that are reassigned or reclassified to another workload.
Working with Workload Monitor Views
The Workload Monitor portlet provides summary views to help you identify issues in workload management.
• Use the Dynamic Pipes view to analyze workload data in near-real time at each system management point of control. You can choose the data sampling period and workload filter criteria. Workloads can be displayed within their enforcement priority (EP).
• Use the Static Pipes view to compare summary and detail workload metrics. Workloads can also be viewed within their EP, and data can be sorted by column. You can also determine whether system resources within an EP are being skewed to a workload by comparing the statistics of a particular workload to other workloads within the EP.
• Use the Distribution view to review workload information within the context of its allocation group (AG). Mouse over the pie charts to see CPU consumption for workloads within an AG compared to the relative weight of the AG. Colors relate pie chart segments to each other and to list information. Use this view to determine:
• Why a workload is over-consuming system resources, causing other workloads to miss their SLGs
• Why a workload is not receiving the system resources required to complete processing
About the State Matrix
The Teradata Database state matrix icon in the toolbar shows changes in state, planned environment, or health condition during the cumulative sampling period.
The state matrix icon uses color to show the following:
Dark blue
Active-state cell.
Medium blue
Previously active state cell
Teradata Viewpoint User Guide, Release 13.11
313
Chapter 30 Workload Monitor
Light blue
Inactive-state cell
Note: During a state change, the cell representing the previous state changes from dark blue to medium blue. If there was a second state change during the sampling period, the previous state cell is shown in light blue.
The number of cells in the state matrix icon depends on the monitored system. If a one-byone state matrix is configured, the state matrix icon appears as one active cell. When you mouse over the state matrix icon, information balloons show detail information and the last state change if there was a state change within the sampling period.
Selecting a Summary View
You can select a summary view. The Workload Monitor portlet saves the active view from your last session to open for your next session.
1
•
•
•
Click one of the following:
to go to the Dynamic Pipes view.
to go to the Static Pipes view.
to go to the Distribution view.
Setting the Sampling Period
The sampling period you select applies to all views that contain a sampling period menu. The
Cumulative System Data menu is dimmed if the selected sampling period has not been reached. The default is 60 minutes. Sampling period totals are updated during the portlet refresh, which is every 60 seconds. Cumulative data is not affected by workload filter criteria.
1
Select a sampling period:
• In the Dynamic Pipes and Static Pipes views under Cumulative System Data from
Last.
• In the Distribution view under CPU Consumption Per Workload from Last.
Workload Monitor Icons
This topic describes the icons used in the Workload Monitor portlet. You can click on the values associated with these icons to see to the details views.
314 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
Icon
Teradata Viewpoint User Guide, Release 13.11
Description
Requests arriving into system management.
Requests that were not allowed to continue processing. This icon appears with the following values:
• The number of rejected requests by system filters
• The number of rejected requests by system throttles
• The number of aborted requests that were caused by exceptions
Requests that have been allowed to continue processing after a warning message has been written to the system log file.
Requests that were delayed and completed or are currently delayed. Does not include requests that were delayed, but are still active.
Requests that use exception processing. This icon appears with these values:
• The total number of workload exceptions in the sampling period
• The number of requests reclassified to a different EP
• The number of requests that have been aborted
315
Chapter 30 Workload Monitor
Icon Description
Completed requests. These requests no longer require system resources.
Navigating to Details Views
Many WORKLOAD MONITOR views provide detailed information and can be accessed from one or more views. To display system data, click the values associated with the arrivals, exceptions, delayed, and completions icons in the Dynamic Pipes, Static Pipes, or workload
name (workload pipe) view.
1
2
3
To display data for an individual workload, click the workload name in any of the toplevel views and do one of the following:
• Click Cumulative to display cumulative values for the selected sampling period.
• Click Snapshot to display current values and the allocation group (workload pipe) view.
• Click Trends 24hrs to display sparkline metrics for the workload.
• Click the allocation group button to display the allocation group details view.
• Click the Cumulative Workload Data to change the sampling period.
To display system data, click the values associated with the arrivals, exceptions, delayed, and completions icons in the Dynamic Pipes, Static Pipes, or workload name (workload pipe) view.
To display workload data and allocation group, do one of the following:
• Click the AG name in the Distribution view.
• Click the AG name button in the workload name (workload pipe) view.
Configuring Columns to Display
Use the Columns dialog box to select, lock, and order columns. You can resize columns in the table.
1
2
3
4
5
6
7
In the table, click .
Click Configure columns.
In the Columns dialog box, select the check boxes of columns to display.
Mouse over the name to see the complete name.
[Optional] Click next to the column name.
The columns at the top of the list can be locked in the table to remain on the left when scrolling horizontally.
[Optional] Click and drag the row to reorder the column.
Click Apply to save changes and close the Columns dialog box.
[Optional] Sort on a column by clicking the column header.
316 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
8
• Sort on a column by clicking the column header. A second click sorts in descending order.
• Sort on two columns consecutively using Ctrl+Click in the column header. Primary sort order is indicated by a single arrow, and secondary sort order is indicated by a double arrow.
• Sort in ascending or descending order on all pages when there are multiple pages.
[Optional] In the table, drag the column heading border in either direction to resize the column.
Workload Monitor Details View Catalog
This topic lists the details views available in the Workload Monitor portlet. Mouse over the error description columns to view the full message.
View Name
ABORTED BY
WORKLOAD
EXCEPTION
ARRIVALS,
COMPLETIONS
CURRENT
ACTIVE
CURRENT
DELAYED
Column and Icon Data
workload, log time, session ID, username, account, error code, reason workload, arrivals, completions, completion %, aborted, delayed, exceptions, met SLG
Accessed from These Views
Dynamic Pipes, Static Pipes,
workload name cumulative and details views
Dynamic Pipes, Static Pipes,
workload name cumulative views
Dynamic Pipes, Static Pipes,
workload name snapshot,
allocation group details views session ID, workload, username, account, CPU time, duration, start, host ID, state, state icon, req CPU, ΔCPU, req I/O, ΔI/O, CPU skew, I/O skew, spool, temp space, PJI, blocked time, in state, partition, CPU use, unnecessary IO, and impact CPU session ID, workload, username, account, blocking count, delay time, system throttle delay time, utility throttle delay time, workload throttle delay time, host ID, duration, start, state, state icon, req CPU, ΔCPU, req I/O, ΔI/O, CPU skew,
I/O skew, spool, temp space, PJI, blocked time, in state, partition, CPU use, unnecessary IO, impact CPU
Note: The system throttle, utility throttle, and workload throttle delay time columns only appear for Teradata Database 13.10 systems.
Dynamic Pipes, Static Pipes,
workload name snapshot,
allocation group details view
EXCEPTIONS summary view
EXCEPTIONS details view workload, exceptions, aborted, reclassified, and logged log time, session ID, username, account, status, rule name, error code, and reason
REJECTED AT
SYSTEM FILTERS log time, session ID, username, account, error code, and reason
Dynamic Pipes and Static
Pipes views
EXCEPTIONS details,
workload name cumulative views
Dynamic Pipes and Static
Pipes views
Teradata Viewpoint User Guide, Release 13.11
317
Chapter 30 Workload Monitor
View Name
REJECTED AT
SYSTEM
THROTTLES and
WORKLOAD
THROTTLES
WARNINGS AT
SYSTEM FILTERS
STATIC PIPES
ENFORCEMENT
PRIORITY
STATIC PIPES
WORKLOADS
Column and Icon Data
log time, session ID, username, account, error code, and reason log time, session ID, username, account, error code, and reason enforcement priority, active, completions, CPU time, met SLG
REQUESTS
DELAYED (active and completed)
RECLASSIFIED
(in and out) workload, enforcement priority, arrivals, completions, aborted, exceptions, met SLG, CPU time, delay (avg), active, delayed, active
(sampling period), and delay queue, reclassified in/out workload, arrivals, delayed, delayed (summary), and delayed (average) log time, session ID, username, into, from, account, status, rule name, error code, reason
Accessed from These Views
Dynamic Pipes and Static
Pipes views
Dynamic Pipes and Static
Pipes views
Static Pipes view
Static Pipes view
Dynamic Pipes and Static
Pipes views
Dynamic Pipes, Static Pipes,
workload name details views
You can link to the views listed in the previous table from the following views:
View Name
workload name
cumulative view
workload name
snapshot view
allocation group
details view
Description
workload, enforcement priority, arrivals, completions, aborted, exceptions, reclassified in/ out workload, enforcement priority, active snapshot, delayed snapshot
Accessed from These Views
Dynamic Pipes, Static Pipes,
Distribution, allocation
group details views
Dynamic Pipes, Static Pipes,
Distribution, allocation
group details views
workload name details views
workload name
trends view workload, enforcement priority, allocation group name, workload name, workload throttle limit, cumulative delayed at workload throttle, percentage of AG CPU for each workload
(sparkline), and percentage of total CPU for the
AG, active requests enforcement priority, sparkline, active queries, average response time, average delay time, maximum response time, delayed queries, average CPU per query, maximum CPU per query, throughput, impact CPU-sec, cumulative
CPU
workload name details views
318 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
About the Dynamic Pipes View
The Dynamic Pipes view helps you analyze workloads. You can access cumulative, snapshot, and trend data for each workload. Use the Filter menu to modify the active workload display. If there are more than 18 workloads, they are grouped by EP.
Use this view to see:
• Workloads that missed their SLGs
• Requests that are delayed or rejected at system filters and system throttles
• Delay queue that is growing. A pulsing indicator appears near the workload name when a workload is adding requests to the delay queue.
Selecting Filter Criteria
1
2
3
4
5
Click Active Workloads.
The Filter dialog box appears.
[Optional] Select the Group by enforcement priority check box to group workloads by their enforcement priority.
[Optional] Select the All Enforcement Priorities check box or click to display the enforcement priority list and select at least one enforcement priority.
[Optional] Select the Missed SLG check box to display only those workloads which have missed their SLGs.
[Optional] Select the Name check box and enter a workload name or name pattern.
Teradata Viewpoint User Guide, Release 13.11
319
Chapter 30 Workload Monitor
6
7
[Optional] Select the Active Workloads check box to display only those workloads which were active within the sampling period.
Click to close the menu.
About the Static Pipes View
Use the Static Pipes view to display workload metrics.
The Workloads tab displays all active workloads within the Cumulative System Data sampling period, the workload EP, and metrics. The following metrics can be added:
• Arrival count
• Aborted count
• Exception count
• Average delay time
• Delayed count
• Active count
The Enforcement Priorities tab displays all active workloads within the Cumulative System
Data sampling period by EP. Available metrics include the total number of active requests classified into one or more workloads in the EP, percentage of completed requests within the
EP, the CPU percentage, and the total number of workloads that met their SLGs.
Click to select the columns displayed on the Workloads tab and the Enforcement
Priorities tab
320 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
About the Distribution View
The Distribution view displays workload CPU consumption percentages, allowing you to compare the CPU consumption for workloads in an AG to the relative weight of the AG within the resource partition.
Use this view to see:
• A pie chart graph that uses color and information balloons to display CPU consumption data for each workload that was active within the Cumulative System Data sampling period. As you mouse over a pie chart section, workloads highlight in the workload list.
Idle CPU time is shown if applicable during the sampling period.
• A pie chart graph that uses color and information balloons to relate the relative weight of each pie section to the AG in the AG list. As you mouse over a pie chart section, the AG highlights in the AG list.
The expand ( ) and collapse ( ) list controls can be used to change the length of the portlet.
Minimizing the lists save space on the portal page.
About the Request Details View
When viewing a request, you can see detailed read-only information from the following tabs:
Overview
Key statistics for a session. Any value exceeding preset thresholds is highlighted.
Teradata Viewpoint User Guide, Release 13.11
321
Chapter 30 Workload Monitor
SQL
SQL for the selected query.
Explain
Explain steps for the query, including step statistics and Explain text.
Blocked By
Details about other queries that are blocking this query.
Delay
Details about rules delaying this query.
Details View Metrics
Query Information
Metric
STATE
TIME IN STATE
Description
Query state, such as ACTIVE, BLOCKED, TERMINATE
How long the query has been in the current state, displayed as
HH:MM:SS
Type
Character
Number
TOTAL DURATION How long the query has run, displayed as HH:MM:SS
CPU USE
Number
Percent of the total amount of available CPU used by the query Percent
CPU
CPU Δ
Total CPU usage time, in seconds, consumed by the query
Total CPU usage time consumed, in seconds, since the last sample
Number
Number
I/O
I/O Δ
CPU SKEW
I/O SKEW
SPOOL
TEMP SPACE
WORKLOAD
PJI
Number of I/O operations the query performed
I/O count since the last sample
CPU skew during the last sample
I/O skew during the last sample
Amount of spool space the query is using
Amount of temp space the query is using
Workload where the query is actively running
Ratio of the CPU milliseconds per I/O for the query, where a larger Product Join Index number indicates system performance degradation
UNNECESSARY I/O All AMP I/O divided by all AMP CPU, displayed in milliseconds, to reveal large amounts of I/O occurring over a short period of time
IMPACT CPU CPU time in seconds of the highest CPU utilized AMP during the collection interval times the total number of AMPs participating for this session during the last session collection interval
Number
Number
Percent
Percent
Number
Number
Character
Number
Number
Number
322 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
Session Information
Metric
USER
ACCOUNT
SOURCE
PARTITION
REQUESTS
Description
Name of the user that submitted the query
Account of the user that submitted the query
Source details, such as application name, IP address, and host user name
Partition in which the query is running
Number of queries submitted by the session
Type
Character
Character
Character
Character
Number
Symbol Reference
The following symbols appear in the portlets:
Symbol Name Definition
ABORTING Query has been aborted, and changes are being rolled back.
ACTIVE
BLOCKED
Query is running.
Query is waiting for a lock held by another query.
DELAYED
IDLE
Query is in a delay queue waiting to run.
No query is running.
OTHER
PARSING
Query is in an unknown state.
Query is being parsed before running.
QTDELAYED Query is waiting for rows to be inserted into a queue table.
RESPONSE Query is returning results to the user.
SESDELAYED Query is delayed because a utility limit has been exceeded in FastLoad,
MultiLoad, FastExport, or ARC.
About the Overview Tab
The Overview tab provides detailed information about key metrics for the selected session and its queries. The metric values displayed provide a snapshot of Teradata Database system status. Metrics that exceed defined thresholds are highlighted.
About the SQL Tab
The SQL tab displays the SQL for the selected query. This information is read-only. Use to export the SQL as a .txt file.
About the Explain Tab
The Explain tab displays an abbreviated version of the Step statistics and the Explain text that result from an Explain request in an SQL session. Each Explain step is uniquely identified with a number where the background of the number box indicates status. This information is read-only. This tab only appears when a query meets certain thresholds established by the Teradata Viewpoint Administrator. Use to export the Explain data as a .csv file.
Teradata Viewpoint User Guide, Release 13.11
323
Chapter 30 Workload Monitor
Step Information
Step Number
Description
• Completed steps are at the top of the list and indicated by a black number box.
• Active steps are indicated by a pulsating number box (flashes gray and white).
• Steps to run are at the bottom of the list and indicated by a white number box.
Confidence Level Indicator Icon
EST. TIME
EST. ROWS
ACTUAL TIME
ACTUAL ROWS
- No confidence in the estimate
- Low confidence in the estimate
- High confidence in the estimate
- High confidence in the estimate due to a join index
Estimated execution time for the step.
Estimated number of rows for the step.
Actual CPU time consumed by the step, or blank if the step has not run.
Actual number of rows for the step, or blank if the step has not run.
About the Blocked By Tab
The Blocked By tab displays details about other queries that are blocking this query. This information is read-only. This tab is only available when the selected query is blocked.
Blocked By Information
USERNAME
HOST
SESSION ID
LOCK TYPE
STATUS
LOCKED
Description
Name of the user that is running the query that holds the lock.
Host ID or LAN ID associated with the PE that processed the logon request for the session.
Session ID of the blocking query.
Type of lock. Type can be Exclusive, Read, Write, or Access.
Lock status. Status can be Waiting or Granted.
Name of the locked object.
About the Delay Tab
The Delay tab displays details about rules that are delaying this query. The Delay tab is available if the system being monitored is a Teradata Database 13.10 system or newer. The information on the Delay tab is read-only, and no actions are available from this tab. This tab is available when the selected query is delayed. The Delay tab displays all rules that apply to the query and a scroll bar appears if there are more than two rules.
Delay Information
BLOCKING COUNT
Description
Number of consecutive times this session has blocked at least one other session.
324 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
Delay Information
UTILITY THROTTLE DELAY
TIME
SYSTEM THROTTLE DELAY
TIME
WORKLOAD THROTTLE
DELAY TIME
RULE NAME
RULE TYPE
OVERRIDABLE
Description
Amount of time request has been delayed by a utility throttle rule.
Amount of time request has been delayed by a system throttle rule.
Amount of time request has been delayed by a workload throttle rule.
Name of rule causing request delay.
TASM type of rule causing request delay.
Indicates if the Teradata Database Administrator can abort or release the request.
About Managing Queries and Sessions
You can manage queries and sessions to improve workload performance and balance system resources.
Abort
Abort the selected query or session.
Change Workload
Change the workload of the selected query or session.
Release Query
Release the selected query from a delay queue.
You must log on with a user ID that has permissions to abort, change workloads, or release queries. If you log out, close a portlet, or open a new portlet instance, you must log on again.
Aborting a Query or a Session
You can abort a query or a session that is blocking other queries or consuming too many resources.
1
2
3
4
5
6
7
From the summary view, click the row of the query you want to abort.
Click .
Select Abort.
[Optional] Log on.
Click Next.
Select one of the following:
• Abort Query to abort the selected query.
• Abort Session to abort the selected query and log off the session.
Click Next.
Teradata Viewpoint User Guide, Release 13.11
325
Chapter 30 Workload Monitor
8
Click Next to confirm your selection.
Changing the Priority of a Query or Session
You can change the priority of a query or session to allow higher priority queries to run or to balance session resources.
This menu item is only available when workloads are not enabled and the system being monitored supports this feature.
1
2
3
From the summary view, click the row of the query with a priority that needs to change.
Click .
Select Change Priority.
4
5
6
7
8
9
10
11
[Optional] Log on.
Click Next.
From the list of accounts that have been assigned to the user, select an account to copy the account name to the Edit new account string box.
[Optional] Edit the account string or enter a new account string.
Click Next.
Select one of the following:
• Move just this query to change the priority of the selected query.
•
Move this query and all future queries associated with this session to change the priority of the selected query and all subsequent queries in the current session.
Click Next.
Click Next to confirm your selection.
Changing the Workload of a Query or Session
You can change the workload of a query or session to either allow higher priority workloads to run or to balance workload resources.
This menu item is only available when workloads are enabled and the system being monitored supports this feature.
1
2
3
4
5
6
7
8
From the summary view, click the row of the query with the workload that needs to change.
Click .
Select Change Workload.
[Optional] Log on.
Click Next.
[Optional] Select a different workload from the list.
Click Next to confirm your selection.
Select one of the following:
326 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
9
10
• Move just this query to change the workload of the selected query.
• Move this query and all future queries associated with this session to change the workload of the selected query and all subsequent queries in the same session.
Click Next.
Click Next to confirm your selection.
Releasing a Query
You can release a query that is in the delay queue and waiting to run.
1
From the summary view, click the row of the query you want to release.
2
3
Click .
Select Release Query.
4
5
6
[Optional] Log on.
Click Next.
Click Next to confirm your selection.
About the Preferences View
The PREFERENCES view allows you to customize the Workload Monitor portlet.
From the portlet frame, click to access the PREFERENCES view and the following tabs:
System and Workloads
Choose a system and its workloads to monitor.
Metrics-Tactical
Select metrics for workloads in the Tactical enforcement priority.
Metrics-Priority
Select metrics for workloads in the Priority enforcement priority.
Metrics-Normal
Select metrics for workloads in the Normal enforcement priority.
Metrics-Background
Select metrics for workloads in the Background enforcement priority.
Trend Interval
Set the time period for calculating each trend interval.
Past Average
Choose the time period for which a historical average is calculated.
Selecting a System and Workloads to Monitor
Use the System and Workloads tab in the PREFERENCES view to select the system and which of its workloads to monitor.
Teradata Viewpoint User Guide, Release 13.11
327
Chapter 30 Workload Monitor
1
2
3
4
5
From the portlet frame, click to access the PREFERENCES view.
Click the System and Workloads tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Select a system from the list.
Select the check box for each workload you want to monitor.
6
7
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Enforcement Priority Metrics
For each enforcement priority group, you can select and order metrics for display from the
PREFERENCES view. Choose from a defined list of metrics.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Metrics tab for an enforcement priority group.
The Preview pane uses sample data to show how metric rows are displayed in the summary view.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Do any of the following to change the way metric rows are displayed:
• Add a metric row. Drag a metric from the Select metrics for display list to the
Preview pane.
• Remove a metric row. Drag a metric from the Preview pane to the Metric Removal pane.
• Change metric row order. Drag a metric already in the Preview pane to a new location in the pane (up or down).
Changes to the metrics affect only the system currently selected, provided that the metrics are available on that system.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting the Workload Trend Interval
Use the Trend Interval tab in the PREFERENCES view to set the length of time used to plot sparkline metrics in the portlet details view.
1
2
From the portlet frame, click to access the PREFERENCES view.
Click the Trend Interval tab.
328 Teradata Viewpoint User Guide, Release 13.11
Chapter 30 Workload Monitor
3
4
5
6
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
In the dialog box, select an Interval.
The default is 24 hours.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Setting Past Averages
Use the Past Averages tab in the PREFERENCES view to specify the number of weeks of data points used to calculate the average values displayed in metric graphs.
1
2
3
4
5
6
From the portlet frame, click to access the PREFERENCES view.
Click the Past Averages tab.
[Optional] Click Clear Defaults to clear all user-defined default settings created for this portlet.
Enter the number of weeks (1 to 99 weeks).
The default is 2 weeks.
[Optional] Select the Use these settings as defaults check box to save the current settings as the default configuration each time you add this portlet to the portal page.
Click OK.
Teradata Viewpoint User Guide, Release 13.11
329
Chapter 30 Workload Monitor
330 Teradata Viewpoint User Guide, Release 13.11
advertisement
Key Features
- Centralized portal for system management
- Variety of portlets for monitoring and management
- Easy to use interface
- Powerful reporting and analysis capabilities
- Scalable and reliable
- Secure and integrated with other Teradata products
